CONFIDENTIAL Second draft for consultation July 2003
Eating Disorders:
Core interventions in the treatment and management of anorexia nervosa, bulimia nervosa, and related eating disorders
National Clinical Practice Guideline Number ___
National Collaborating Centre for Mental Health Commissioned by the National Institute for Clinical Excellence
CONFIDENTIAL Second draft for consultation September 2003
Eating Disorders Guideline Development Group Members
Professor Simon Gowers Professor of Adolescent Psychiatry University of Liverpool Guideline Development Chair
Mr Stephen Pilling Co-Director, National Collaborating Centre for Mental Health, and Camden and Islington Mental Health and Social Care Trust.
Professor Janet Treasure Professor of Psychiatry Eating Disorders Unit Guys Kings and St Thomas School of Medicine Kings College London South London and Maudsley NHS Trust Lead, Topic Group on Physical Management
Professor Christopher Fairburn Wellcome Principal Research Fellow and Professor of Psychiatry Department of Psychiatry, Oxford University Lead, Topic Group on Psychological Interventions
Dr Bob Palmer Senior Lecturer in Psychiatry University of Leicester Lead, Topic Group on Service-level Interventions
Dr Lorraine Bell Consultant Clinical Psychologist Eating Disorders Team, Portsmouth Health Care NHS Trust
Ms Nicky Bryant Chief Executive Eating Disorders Association (March 2002-2003)
Dr Rachel Bryant-Waugh Consultant Clinical Psychologist, West Hampshire NHS Trust Honorary Senior Lecturer, University of Southampton
Mr Peter Honig Family Therapist Phoenix Centre Eating Disorders Service Cambridgeshire and Peterborough Mental Health Partnership NHS Trust
2 CONFIDENTIAL Second draft for consultation September 2003
Dr Pippa Hugo Child and Adolescent Psychiatrist St. George's Eating Disorder Service, South West London and St. George's Mental Health NHS Trust.
Dr Robert Mayer General Practitioner Highgate Group Practice, London
Mr Ciaran Newell Consultant Nurse Eating Disorder Service, Dorset Healthcare NHS Trust
Ms Jane Nodder Patient representative, London
Dr Deborah Waller General Practitioner 19 Beaumont Street, Oxford
Ms Susan Ringwood Chief Executive Eating Disorders Association (December 2002 – January 2004)
Dr Ulrike Schmidt Senior Lecturer in Eating Disorders Eating Disorders Unit, Institute of Psychiatry
NCCMH Staff
Dr Catherine Pettinari: Senior Project Manager Dr Craig Whittington: Senior Systematic Reviewer Dr Judit Simon: Health Economist Ms Heather Wilder: Information Scientist Ms Ellen Boddington: Research Assistant Mr Lawrence Howells: Research Assistant
3 CONFIDENTIAL Second draft for consultation September 2003
1 Introduction 7
1.1 National guidelines 8 1.2 The national eating disorders guideline 11
2 Eating disorders 14
2.1 Anorexia nervosa 14 2.2 Bulimia nervosa 19 2.3 Atypical eating disorders (eating disorders not otherwise specified; EDNOS) including binge eating disorder 23 2.4 Incidence and prevalence of eating disorders 25 2.5 The aetiology of eating disorders 27 2.6 Use of health service resources 30 2.7 The treatment and management of eating disorders in the NHS 34 2.8 Assessment 39 2.9 Engagement, consent and the therapeutic alliance 41 2.10 The patient experience of eating disorders 42 2.11 The involvement of family members / carers 46 2.12 Stigma 50
3 Methods used to develop this guideline 52
3.1 Overview 52 3.2 The Guideline Development Group 52 3.3 Clinical questions 54 3.4 Systematic clinical literature review strategies 54 3.5 Health economics review strategies 66 3.6 Stakeholder contributions 66 3.7 Validation of this guideline 67
4 Summary of recommendations 68
4.1 Introduction 69 4.2 Care across all conditions 69 4.3 Identification and management of eating disorders in primary care and non-mental health settings 70 4.4 Anorexia nervosa 71 4.5 Bulimia nervosa 77
4 CONFIDENTIAL Second draft for consultation September 2003
4.6 Treatment and management of atypical eating disorders including binge eating disorder 80 4.7 Research recommendations 81
5 Identification of eating disorders in primary care 83
5.1 Introduction 83 5.2 Screening 84
6 Treatment and management of anorexia nervosa 91
6.1 Introduction 91 6.2 Psychological interventions 91 6.3 Pharmacological interventions 104 6.4 Management of physical aspects 107 6.5 Service interventions 120 6.6 Predicting the outcome of treatment and recovery from anorexia nervosa 131
7 Treatment and management of bulimia nervosa 133
7.1 Introduction 133 7.2 Psychological interventions 133 7.3 Pharmacological interventions 144 7.4 Antidepressant drugs compared to psychological interventions and their combination 150 7.5 Management of the physical aspects of bulimia nervosa 159 7.6 Service level interventions 168 7.7 Predicting the outcome of treatment and recovery from bulimia nervosa 170
8 Treatment and management of atypical eating disorders (eating disorders not otherwise specified) including binge eating disorder 173
8.1 Introduction 173 8.2 Psychological interventions 173 8.3 Pharmacological interventions 181 8.4 Management of physical aspects 183 8.5 Service level interventions 183
9 Health economics evidence 184
9.1 Systematic literature review 184 9.2 Cost-effectiveness modelling 188
5 CONFIDENTIAL Second draft for consultation September 2003
10 Criteria for auditing the management of eating disorders 200
10.1 Objectives for the audit 200 10.2 Individuals to be included in an audit 200 10.3 Measures that could be used as a basis for an audit 200
11 References 209
12 Glossary 238
6 CONFIDENTIAL Second draft for consultation September 2003
1 Introduction
This guideline has been developed to advise on the identification, treatment and management of the eating disorders anorexia nervosa, bulimia nervosa and related conditions. The guideline recommendations have been developed by a multidisciplinary group of healthcare professionals, patients and their representatives, and guideline methodologists after careful consideration of the best available evidence. It is intended that the guideline will be useful to clinicians and service commissioners in providing and planning high quality care for those with eating disorders while also emphasising the importance of the experience of care for patients and carers.
Eating disorders comprise a range of syndromes encompassing physical, psychological and social features. Whilst the acute physical complications of these disorders may provoke great concern in family members and health service staff, anorexia nervosa and bulimia nervosa are frequently chronic conditions with substantial long-term physical and social sequelae, from which recovery is difficult. Long-term disabilities include negative effects on employment, fertility, relationships and parenting. The impact of a person’s eating disorder on home and family life is often considerable and family members may carry a heavy burden over a long period of time. Family members are often at a loss to know how to help and offer support to an affected relative.
About 1 in 250 females and 1 in 2000 males will experience anorexia nervosa, generally in adolescence or young adulthood. About five times that number will suffer from bulimia nervosa. Other atypical eating disorders are more common still, though many will not receive treatment. As eating disorders commonly develop during adolescence, they can blight physical and social development and many sufferers fail to reach their academic potential. Depressed mood is a common feature, partly because of these adverse consequences and also because of the distressing nature of the central symptoms of these disorders. The adverse physical consequences of dieting, weight loss and purging behaviours are notable and sometimes prove fatal. Indeed, anorexia nervosa has the highest mortality rate of any psychiatric disorder of adolescence.
The treatment experience of those with eating disorders is extremely variable. In part, this relates to the inherent ambivalence to treatment commonly experienced by those with these conditions. It is also due to the uneven provision of effective psychiatric treatments that range from high quality age- appropriate specialist services, to basic generic provision in areas of the country where skills and experience are scarce. Sadly, a number of those with eating disorders will receive negative attitudes from inexperienced clinical
7 CONFIDENTIAL Second draft for consultation September 2003
staff and they may on occasion fear being trapped in treatment rather than helped by it.
This guideline addresses aspects of service provision, physical management and therapeutic approaches for those with eating disorders from the age of 8 upwards. Although the evidence base is rapidly expanding, there are a number of major gaps and future revisions of this guideline will incorporate new scientific evidence as it develops. The guideline makes a number of research recommendations specifically to address these gaps in the evidence base. In the meantime, we hope that the guideline will assist clinicians, patients and carers by identifying the merits of particular treatment approaches where the evidence from research and clinical experience exists.
1.1 National guidelines
1.1.1 What are clinical practice guidelines? Clinical practice guidelines are ‘systematically developed statements that assist clinicians and patients in making decisions about appropriate treatment for specific conditions’ (Mann, 1996). They are derived from the best available research evidence, using predetermined and systematic methods to identify and evaluate the evidence relating to the specific condition in question. Where evidence is lacking, the guidelines will incorporate statements and recommendations based upon the consensus statements developed by the guideline development group.
Clinical guidelines are intended to improve the process and outcomes of health care in a number of different ways. Clinical guidelines can:
• provide up-to-date evidence-based recommendations for the management of conditions and disorders by healthcare professionals
• be used as the basis to set standards to assess the practice of healthcare professionals
• form the basis for education and training of healthcare professionals
• assist patients and carers in making informed decisions about their treatment and care
• improve communication between healthcare professionals, patients and carers
• help identify priority areas for further research.
8 CONFIDENTIAL Second draft for consultation September 2003
1.1.2 Uses and limitations of clinical guidelines Guidelines are not a substitute for professional knowledge and clinical judgment. Guidelines can be limited in their usefulness and applicability by a number of different factors: the availability of high quality research evidence, the quality of the methodology used in the development of the guideline, the generalisability of research findings and the uniqueness of individual patients.
Although the quality of research in eating disorders is variable, the methodology used here reflects current international understanding on the appropriate practice for guideline development (AGREE: Appraisal of Guidelines for Research and Evaluation Instrument; www.agreecollaboration.org), ensuring the collection and selection of the best research evidence available, and the systematic generation of treatment recommendations applicable to the majority of patients and situations. However, there will always be some people and situations for which clinical guideline recommendations are not readily applicable. This guideline does not, therefore, override the individual responsibility of healthcare professionals to make appropriate decisions in the circumstances of the individual patient, in consultation with the patient and/or carer.
In addition to the clinical evidence, cost-effectiveness information, where available, is taken into account in the generation of statements and recommendations of the clinical guidelines. While national guidelines are concerned with clinical and cost effectiveness, issues of affordability and implementation costs are to be determined by the NHS.
In using guidelines, it is important to remember that the absence of empirical evidence for the effectiveness of a particular intervention is not the same as evidence for ineffectiveness. In addition, of particular relevance in mental health, evidence-based treatments are often delivered within the context of an overall treatment programme including a range of activities, the purpose of which may be to help engage the patient, and provide an appropriate context for the delivery of specific interventions. It is important to maintain and enhance the service context in which these interventions are delivered; otherwise the specific benefits of effective interventions will be lost. Indeed, the importance of organising care, so as to support and encourage a good therapeutic relationship, is at times as important than the specific treatments offered.
1.1.3 Why develop national guidelines? The National Institute for Clinical Excellence (NICE) was established as a Special Health Authority for England and Wales in 1999, with a remit to provide a single source of authoritative and reliable guidance for patients, professionals and the public. NICE guidance aims to improve standards of care, to diminish unacceptable variations in the provision and quality of care
9 CONFIDENTIAL Second draft for consultation September 2003 across the NHS and to ensure that the health service is patient-centred. All guidance is developed in a transparent and collaborative manner using the best available evidence and involving all relevant stakeholders.
NICE generates guidance in a number of different ways, two of which are relevant here. First, national guidance is produced by the Technology Appraisal Committee to give robust advice about a particular treatment, intervention, procedure or other health technology. Second, NICE commissions the production of national clinical practice guidelines focused upon the overall treatment and management of a specific condition. To enable this latter development, NICE has established seven National Collaborating Centres in conjunction with a range of professional organisations involved in healthcare.
1.1.4 The National Collaborating Centre for Mental Health This guideline has been commissioned by NICE and developed within the National Collaborating Centre for Mental Health (NCCMH). The NCCMH is a collaboration of the professional organisations involved in the field of mental health, national patient and carer organisations, a number of academic institutions and NICE. The NCCMH is funded by NICE and is led by a partnership between the Royal College of Psychiatrists’ research unit (College Research Unit – CRU) and the British Psychological Society’s equivalent unit (Centre for Outcomes Research and Effectiveness – CORE). Members of the NCCMH reference group come from the following organisations:
• Royal College of Psychiatrists (RCPsych)
• British Psychological Society (BPS)
• Royal College of Nursing (RCN)
• Social Care Institute of Excellence (SCIE)
• College of Occupational Therapists (COT), now replaced by the Clinical Effectiveness Forum for the Allied Health Professions (CEFAHP)
• Royal College of General Practitioners (RCGP)
• Royal Pharmaceutical Society (RPS)
• Rethink Severe Mental Illness
• Manic Depression Fellowship (MDF)
10 CONFIDENTIAL Second draft for consultation September 2003
• MIND
• Centre for Evidence Based Mental Health (CEBMH)
• Centre for the Economics of Mental Health (CEMH)
• Institute of Psychiatry (IoP).
The NCCMH reference group provide advice on a full range of issues relating to the development of guidelines, including the membership of experts, professionals, patients and carers within guideline development groups.
1.1.5 From national guidelines to local protocols Once a national guideline has been published and disseminated, local healthcare groups will be expected to produce a plan and identify resources for implementation, along with appropriate timetables. Subsequently, a multidisciplinary group involving commissioners of health care, primary care and specialist mental healthcare professionals, patients and carers should undertake the translation of the implementation plan into local protocols. The nature and pace of the local plan will reflect local health care needs and the nature of existing services; full implementation may take a considerable time, especially where substantial training needs are identified.
1.1.6 Auditing the implementation of guidelines This guideline identifies key areas of clinical practice and service delivery for local and national audit. Although the generation of audit standards is an important and necessary step in the implementation of this guidance, a more broadly based implementation strategy will be developed. Nevertheless, it should be noted that the Commission for Healthcare Audit and Inspection (CHAI) will monitor the extent to which Primary Care Trusts (PCTs) and specialist secondary care trusts responsible for mental health and social care and Health Authorities have implemented these guidelines.
1.2 The national eating disorders guideline
1.2.1 Who has developed this guideline? The Guideline Development Group (GDG) was convened by the NCCMH based upon advice from the Centre’s reference group representatives, and supported by funding from NICE. The GDG included members from the following professional groups: psychiatry, clinical psychology, nursing,
11 CONFIDENTIAL Second draft for consultation September 2003
family therapy, social work and general practice. In addition, the GDG included a patient1 and a representative of The Eating Disorders Association.
Staff from the NCCMH provided leadership and support throughout the process of guideline development, undertaking systematic searches, information retrieval, appraisal and systematic review of the evidence. Members of the GDG received training in the process of guideline development from the Centre for Evidence-Based Mental Health (CEBMH), and support from the NICE Patient Involvement Unit. The NICE Guidelines Technical Advisor provided advice and assistance regarding aspects of the guideline development process.
All GDG members made formal declarations of interest at the outset, which were updated at every GDG meeting. The GDG met a total of 23 times throughout the process of guideline development. The GDG formed sub- groups, or “Topic Groups” for ease of evidence identification and analysis and to address identifiable treatment approaches. Each Topic Group was led by a national expert in the relevant topic and the groups supported by the NCCMH technical team, with additional expert advice from special advisors where necessary. Topic Groups oversaw the production and synthesis of research evidence before presentation to the wider GDG. All statements and recommendations in this guideline have been generated and agreed by the whole GDG.
1.2.2 For whom is this guideline intended? This guideline will be of relevance to all people with a diagnosis of anorexia nervosa, bulimia nervosa or related eating disorders aged 8 years of age and over. The guideline will not explicitly provide guidance on the diagnosis or treatment of people with eating disorders if there is a separate physical or other primary mental disorder of which a disorder of eating is a symptom. This may be dealt with in a future guideline.
The guideline will review the issue of diagnosis and assessment, as many people suffer with eating disorders that fall outside established diagnostic criteria.
In sum, this guideline is intended for use by:
• Individuals with a diagnosis of anorexia nervosa, bulimia nervosa or related conditions aged 8 years and over and their families/carers
• Professional groups who share in the treatment and care for people with a diagnosis of an eating disorder, including psychiatrists, clinical psychologists, mental health nurses, community psychiatric nurses,
1 The term “patient” was the preferred term for use in the guideline based on a survey of Eating Disorders Association members.
12 CONFIDENTIAL Second draft for consultation September 2003
social workers, practice nurses, secondary care medical, dietitians, dental, nursing and paramedical staff, occupational therapists, pharmacists, paediatricians, other physicians, general medical and dental practitioners, physiotherapists and family/other therapists
• Professionals in other health and non-health sectors who may have direct contact with or are involved in the provision of health and other public services for those diagnosed with eating disorders. These may prison doctors, the police and professionals who work in the criminal justice and education sectors
• Those with responsibility for planning services for people with a diagnosis of an eating disorder and their carers, including directors of public health, NHS trust managers and managers in PCTs.
1.2.3 Specific aims of this guideline The guideline makes recommendations for the identification, treatment and management of eating disorders. Specifically, it aims to:
• Evaluate the role of specific psychological interventions in the treatment and management of eating disorders
• Evaluate the physical management and role of specific pharmacological agents in the treatment of eating disorders
• Evaluate the role of specific service delivery systems and service-level interventions in the management of eating disorders
• Integrate the above to provide best practice advice on the care of individuals with a diagnosis of eating disorders through the different phases of illness, including the initiation of treatment, the treatment of acute episodes and the promotion of recovery.
13 CONFIDENTIAL Second draft for consultation September 2003
2 Eating disorders
This guideline is concerned with the identification, treatment and management of anorexia nervosa and bulimia nervosa as defined in the 10th Edition of the International Classification of Diseases (ICD 10) (WHO, 1992). The guideline does not address the management of loss of appetite, psychogenic disturbance of appetite or other conditions that involve significant weight loss but which are due to known physical illness. The guideline is also concerned with other related disorders that do not fulfil diagnostic criteria for anorexia nervosa or bulimia nervosa. These are generally called ‘Atypical Eating Disorders’ (Fairburn & Harrison, 2003) or Eating Disorder Not Otherwise Specified (EDNOS) (APA, 1994). The American Psychiatric Association’s Diagnostic and Statistical Manual 4th Revision further describes the diagnostic category Binge Eating Disorder as a research diagnosis within EDNOS.
2.1 Anorexia nervosa
2.1.1 Symptoms, presentation and patterns of illness Anorexia nervosa is a syndrome in which the individual maintains a low weight as a result of a preoccupation with body weight, construed either as a fear of fatness or pursuit of thinness. In anorexia nervosa, weight is maintained at least 15% below that expected, or in adults Body Mass Index (BMI) - calculated as weight in kilograms divided by height in metres squared - is below 17.5 kg/m2. In younger people, the diagnosis may be made in those who fail to gain weight during the expected growth spurt of puberty, as they can become underweight without weight loss.
Weight loss in anorexia nervosa is induced by avoiding “fattening foods”, sometimes supported by excessive exercising or self-induced purging (by vomiting or misuse of laxatives). As a consequence of poor nutrition, a widespread endocrine disorder involving the hypothalamic-pituitary-gonadal axis develops, manifest in women by amenorrhoea and in men by a lack of sexual interest or potency. In prepubertal children, puberty is delayed and growth and physical development are usually stunted.
The subjective experience of anorexia nervosa is often at odds with the assessment of others. The conviction that weight control is desirable is usually strongly held, particularly when challenged and others are seen as mistaken in believing the person should gain weight, particularly where there is a marked disturbance of body image. Weight loss is experienced as a positive achievement and therefore may be strongly reinforcing to someone with low confidence and poor self-esteem. As a result, they will often deny the
14 CONFIDENTIAL Second draft for consultation September 2003
seriousness of the condition. The essential role of ‘weight phobia’ is increasingly being questioned however, and is believed by some to be culture specific.
The condition generally starts with dieting behaviour that may evoke no concern. Indeed, some will experience reinforcing compliments. After a while however, the commitment to dieting increases, often with a number of secondary features such as social withdrawal, rigidity and obsessionality, particularly where these traits have previously been features of the person’s personality. A number of secondary handicaps may develop including physical adverse effects, social isolation, compromise of educational or occupational plans and potential in the areas of leisure, self care, daily living and productivity of employment and/or education. A smaller number will enter anorexia nervosa through a pattern of purging behaviour without dieting, following a viral illness, which resulted in weight loss that then became positively valued, or in the context of a chronic illness such as diabetes or Crohns disease.
Typically individuals are persuaded to seek help by concerned family members, teaching staff or general practitioners with whom they consult about physical consequences. Sometimes, however, the person begins to appreciate the damaging effects of the disorder and may seek help in their own right. Children and adolescents are almost always brought to treatment, very rarely actively seeking help initially.
2.1.2 Diagnosis The diagnosis of anorexia nervosa in its typical form is a relatively straightforward one in older adolescents and adults. The diagnosis has good validity and reliability, the main obstacle to diagnosis being the person’s own willingness or otherwise to disclose his or her motives, symptoms and behaviours. Thus, engagement in a supportive, empathic assessment interview is crucial in enabling the person to reveal fears around weight, dieting behaviour and any purging or other maladaptive behaviour such as excessive exercising. In the absence of this engagement, the individual may fail to reveal weight-controlling behaviours and collude with the doctor in pursuing physical investigations to explain the weight loss. In women, the presence of secondary amenorrhoea (i.e., cessation of menstruation after it has been established) or other physical features of starvation should always alert the physician to the possibility of this diagnosis. Diagnosis may be more problematic in children and younger adolescents, as the existing diagnostic criteria are insufficiently developmentally sensitive.
The diagnosis is made on the basis of the history, supported where possible by a corroborative account from a relative or friend. Physical examination, with measurement of weight and height and calculation of Body Mass Index (BMI), can reveal the extent of emaciation. On occasion, clinical observation
15 CONFIDENTIAL Second draft for consultation September 2003 during a hospital assessment can enable characteristic behaviours to be observed. Physical investigations are less useful in making the diagnosis but are crucial in assessing the physical impact of the disorder and its complications. Depending on the results of the physical examination, these may include haematological tests, electrocardiography, radiological assessment and ultrasound (Royal College of Psychiatrists, 2002).
A diagnostic challenge occurs in those with co-morbid physical disorders, such as diabetes, chronic bowel or thyroid disorder. In diabetes, the patient may be tempted to restrict insulin intake in order to lose calories, whilst on occasions the symptoms of organic intestinal disorder may mask the psychological condition.
The weight loss that occurs with the anorexia of depression can usually be distinguished from that resulting from the dietary control of anorexia nervosa, but the condition can sometimes be difficult to distinguish from post viral and other chronic fatigue syndromes where food intake is poor. Weight loss and limited food intake secondary to a brain tumour is also known to have been mistaken for anorexia nervosa.
2.1.3 Physical and social consequences Although in the acute stages of anorexia nervosa, subjective distress may be limited, emotional disturbance is common, chiefly comprising anxiety and mood symptoms. With time, emotional difficulties usually increase along with a range of physical and social difficulties, including becoming unable to care for oneself adequately, reducing or stopping leisure activities, interrupting educational goals and losing personal autonomy. These affect the person’s quality of life and increase the reliance on and the importance of the eating disorder.
Depression is a common co-morbid diagnosis, with rates of up to 63% in some studies (Herzog et al., 1992), while obsessive-compulsive disorder (OCD) has been found to be present in 35% of patients with anorexia nervosa (Rastam, 1992).
Physical problems can be classified as those due to the effects of starvation and the consequences of purging behaviour. Starvation affects every system in the body. In the musculo-skeletal system, this will be evident as weakness, loss of muscle strength (which also affects heart muscle), loss of bone density and impairment of linear growth. Young women with anorexia nervosa are at increased risk of bone fractures later in life (Lucas et al., 1999). The effects on the endocrine system have their impact on target organs, causing infertility, a risk of polycystic ovaries and loss of bone mineralisation. Where pubertal development has not been completed, incomplete development of secondary sexual characteristics may occur (Goldbloom & Kennedy, 1995) and permanent stunting of growth is common. The effects of purging are
16 CONFIDENTIAL Second draft for consultation September 2003 described in section 7.5.2, including long-term disabilities such as erosion of tooth enamel sometimes amounting to destruction of the whole dentition. Worn painful teeth can be a considerable concern to the patient in terms of comfort, appearance and therefore self-esteem.
Brain volume is reduced in anorexia nervosa (Dolan, Mitchell & Wakeling, 1988; Kohn et al., 1997; Kingston, Szmukler, Andrewes, Tress & Desmond, 1996; Krieg, Pirke, Lauer & Backmund, 1988; Swayze et al., 1996). There are two small longitudinal studies, which have examined the structural changes in the brain of adolescents after full weight gain (Golden et al., 1996; Katzman et al., 1996). Both found persistent deficits in grey matter (cell bodies of neurons and glial cells) although there was recovery of white matter (mainly myelinated axons). This supports the finding of grey matter deficits in people who have made a full recovery from their eating disorder (Lambe, Katzman, Mikulis, Kennedy & Zipursky, 1997). One post mortem study reported that there was a reduction in basal dendritic fields and dendritic spine density (Neumarker et al., 1997).
Many of the cognitive deficits in anorexia nervosa are restored after weight recovery. However, some abnormalities in executive function remain after weight restoration. For example, people with eating disorders have scores greater than one standard deviation from the norms on tests of perceptual rigidity, perseveration and set shifting and the neurological sign dysdiadokinesis (Tchanturia, Morris, Surguladze & Treasure, 2002). Although little is known of the effects short or long term of extreme weight loss on brain development and function in children, it is possible that such weight loss may have both short and long-term effects on cognitive functioning.
Social difficulties may result in continued dependence on family of origin into adult life and often include difficulties engaging in intimate relationships. Employment prospects may be adversely affected either because of the limitations of the disorder or the disruption caused by lengthy hospitalisations.
2.1.4 Course and prognosis The course of anorexia nervosa is very variable. There is no good evidence on the prognosis for people with anorexia nervosa who do not access formal medical care (Treasure & Schmidt, 2002). A summary of 68 treatment studies published before 1989 with a length of follow-up of 1 to 33 years, found that 43% of people recover completely, 36% improve, 20% develop a chronic eating disorder and 5% die from anorexia nervosa, (Steinhausen, 1995). The overall mortality in these long-term studies ranged from 0-21% from a combination of physical complications and suicide. The all cause standardised mortality ratio anorexia nervosa has been estimated at 9.6 (95% Confidence Interval 7.8 to 11.5) Nielson (2001), about three times higher than other psychiatric illnesses. The average annual risk of mortality has been calculated
17 CONFIDENTIAL Second draft for consultation September 2003 at 0.59% per year in females averaged from 10 samples, with a minimum follow-up of six years (Neilsen et al., 1998). The mortality rate appears to be higher for people with lower weight during their illness and those presenting between 20 and 29 years of age.
A number of those with anorexia nervosa progress to other eating disorders, particularly bulimia nervosa, but also binge eating disorder, highlighting the relationship between the disorders. Movement in the other direction is less common, but a number of those with anorexia nervosa give a premorbid history of obesity in childhood or adolescence.
2.1.5 Anorexia nervosa in children and adolescents Although the essential psychological features are similar, children and younger adolescents may present with delayed puberty or stunted growth as well as weight loss. Parents or teachers are generally the ones who raise concern and the young person may resist medical attention. Some young people will voice anxieties around unwanted aspects of development, particularly if they have experienced early puberty or feel unable to engage with their peers’ increasing adolescent independence and social experimentation. In some, bullying or teasing about weight may have provoked this concern.
Although the principles of making the diagnosis are the same as in adults and are often straightforward, the greatest diagnostic difficulty occurs in the youngest cases. In children between the ages of around 8 and 12, the condition is less common than in older individuals and should be distinguished from other types of eating disturbance seen in middle childhood, such as selective eating and food avoidance emotional disorder. By definition feeding disorder of infancy and childhood has onset below age 6. In pubescent cases with primary amenorrhoea, it can sometimes be difficult to judge whether puberty has been delayed, from the normal variation in timing of puberty. Reference to height and weight centile charts is useful in evaluating weight in comparison to height. It is particularly helpful to compare presenting centiles for weight and height with historical values, as these may identify stunting of height (where the young person has crossed height centile lines). The result of such stunting is that the person may not appear unduly thin, though his or her weight may be considerably below the projected level as indicated by premorbid height and weight. It is also helpful to plot body mass index on BMI centile charts, as BMI norms are not stable over age. Average BMI increases with age during childhood and adolescence, a BMI of 17.5 kg/m2 being close to the mean for a child at the age of 12 (Cole et al., 1995).
The prognosis for children and adolescents with anorexia nervosa is variable. Some (particularly those with a rapid and early onset) will make a full recovery from a first episode. This is most likely where early physical and
18 CONFIDENTIAL Second draft for consultation September 2003
psychosocial development has been healthy and where there is an identified precipitating negative life event such as bereavement (North et al., 1997). In such cases and where onset is pre-pubertal, physical consequences such as stunted growth and pubertal delay are usually fully reversible. Others with a more insidious onset, with earlier social difficulties or abnormal personality development, may go on to have a more chronic course into middle age (Gowers et al., 1991).
2.2 Bulimia nervosa
2.2.1 Symptoms, presentation & pattern of illness Bulimia nervosa is characterised by recurrent episodes of binge eating and secondly by compensatory behaviour (vomiting, purging, fasting or exercising) in order to prevent weight gain. Binge eating is accompanied by a subjective feeling of loss of control over eating. Self-induced vomiting and excessive exercise, as well as the misuse of laxatives, diuretics, thyroxine, amphetamine or other medication, may occur. As in anorexia nervosa, self- evaluation is unduly influenced by body shape and weight, and there may indeed have been an earlier episode of anorexia nervosa. The diagnosis of anorexia nervosa is given precedence over bulimia nervosa; hence in bulimia nervosa BMI is maintained above 17.5 kg/m2 in adults and the equivalent in children and adolescents (see section 2.1.5). There is some controversy concerning whether those who binge eat but do not purge should be included within this diagnostic category. The ICD10 criteria (WHO, 1992) stress the importance of purging behaviour on the grounds that vomiting and laxative misuse are considered pathological behaviours in our society in comparison to dieting and exercise. The DSM-IV criteria (APA, 1994), agree about the importance of compensatory behaviour but distinguish between the purging type of bulimia nervosa in which the person regularly engages in self induced vomiting or the misuse of laxatives, diuretics or enemas, from the non- purging type in which other inappropriate compensatory behaviours such as fasting or excessive exercise occur but not vomiting or laxative misuse.
People with bulimia nervosa tend to not disclose their behaviour nor to seek out treatment readily although may be more likely to do so than those with anorexia nervosa. The condition appears to be subjectively less ‘valued’ than anorexia nervosa; indeed binge eating and purging are commonly associated with extreme subjective guilt and shame. These emotions are sometimes reinforced by the pejorative language used by relatives and others including sometimes clinicians, who may refer to “confessing” or “admitting” to purging behaviour. A person’s ambivalence towards treatment often arises from the fear that they will be stopped from vomiting and purging and then left to face the consequences of their binge eating, i.e., excessive weight gain. The condition usually develops at a slightly older age than anorexia nervosa (the mean age of onset is 18 to 19, compared to 16 to 17 for anorexia nervosa).
19 CONFIDENTIAL Second draft for consultation September 2003
Bulimia nervosa sometimes arises from a pre-existing anorexic illness. Where this is not the case the development of the disturbance is often essentially similar to that of anorexia nervosa, arising from a background of attempts to restrain eating. In bulimia nervosa however, dietary restriction cannot be maintained and is broken by episodes of reactive binge eating which result from a combination of physiological and psychological factors. Compensatory behaviours follow in order to counteract the effect of binge eating on weight. The person therefore maintains a weight, usually within the normal range despite overeating but commonly progresses into a vicious cycle of attempted dieting, binge eating and compensatory purging, frequently on a daily basis. As these behaviours dominate daily life, the person becomes preoccupied with thoughts of food and life may be re-organised around shopping, eating and purging behaviour. Initially, those with bulimia nervosa are generally secretive about their bulimic episodes, though some may leave obvious signs of their disorder such as empty food packaging and occasionally bags of vomit for other family members to discover.
Bulimic episodes are frequently planned, with food purchased or prepared in order to be consumed without interruption. The individual may also avoid situations in which they are likely to be exposed to food or will find it difficult to control their eating, such as when eating out with others. This avoidance behaviour tends to add to any social and relationship difficulties that may be present.
Mood disturbance is extremely common in bulimia nervosa and symptoms of anxiety and tension are frequently experienced. Self-denigratory thoughts may develop out of disgust at overeating or purging whilst low self esteem and physical self-loathing may in some be rooted in the past experience of physical or sexual abuse. Self-harm, commonly by scratching or cutting is common. A significant proportion of those with bulimia nervosa have a history of disturbed interpersonal relationships with poor impulse control. Some will abuse alcohol and drugs.
2.2.2 Diagnosis As in anorexia nervosa, the diagnosis depends on obtaining a history supported as appropriate by the corroborative account of a parent or relative. This will require an empathic supportive non-judgemental interview style in which the person is enabled to reveal the extent of their symptoms and behaviours. Although those with bulimia nervosa generally have fewer serious physical complications than those with anorexia nervosa, they commonly report more physical complaints when first seen. They may complain of fatigue, lethargy, or feeling bloated, and they may suffer constipation, abdominal pain and on occasions swelling of the hands and feet or irregular menstruation. Erosion of dental enamel in which the lingual surface of the upper teeth is mainly affected, is virtually pathognomonic of vomiting (Mitchell, 1995).
20 CONFIDENTIAL Second draft for consultation September 2003
Physical examination is often normal, though the salivary glands (particularly the parotid glands) may be enlarged. Calluses on the back of the hand may be found; these result from the use of the hand to stimulate the gag reflex and induce vomiting. Oedema is common in those who have used laxatives or diuretics whilst these behaviours also lead to fluid loss and subsequent dehydration, which in turn may result in a metabolic alkalosis. This is generally accompanied by hypochloraemia and hypokalaemia. Overall, about 10% of those with bulimia nervosa have electrolyte abnormalities detected on routine screening (Mitchell, 1995). Metabolic acidosis can also occur in patients who are abusing laxatives as a result of the loss of bicarbonate from the bowel. Less commonly hyponatraemia, hypocalcaemia, hypophosphataemia and hypomagnesaemia may develop.
In the gastrointestinal system, oesophagitis may occur. Gastric dilatation that poses the risk of gastric rupture and death occurs rarely but may be the most common cause of fatality (Mitchell, 1997). Constipation is extremely common, mainly due to dehydration; steatorrhoea and protein losing gastroenteropathy have also been reported secondary to laxative abuse. Severe use of laxatives has been reported to cause cardiomyopathy and other types of myopathic disorder as a result of storage of the Ipecac in muscle tissue (Mitchell, 1995). Electrocardiography may indicate heart conduction abnormalities and possibly an increase in the risk of mitral valve prolapse.
Abnormal electroencephalographic (EEG) findings have been reported probably secondary to fluid and electrolyte abnormalities; there does not seem to be an association between bulimia nervosa and epilepsy.
Endocrine abnormalities are variable. These include abnormalities in the menstrual cycle and blunting of the thyroid stimulating hormone and growth hormone response to thyroid releasing hormone.
2.2.3 Impairment and disability There is considerable overlap between the long term disabling consequences of bulimia nervosa and those of anorexia nervosa. Mood and anxiety symptoms are very common. These symptoms, low self esteem and body image disturbance can all have a negative effect on social relationships, which in turn may be damaged by a lifestyle that may be chaotic and characterised by impulsivity.
The adverse physical consequences of purging behaviour have been identified in Section 7.5.2. In addition, those with bulimia nervosa may be at risk from the effects of alternating weight loss and weight gain.
Lissner et al. (1991) in a series of 3,130 participants found that both all-cause and coronary heart disease mortality were increased significantly in both men
21 CONFIDENTIAL Second draft for consultation September 2003 and women with high levels of weight variability. Morbidity from coronary heart disease was also increased.
Estimates of the prevalence of the diagnoses of personality disorder in people with bulimia nervosa have ranged from 21% to 77%. Obsessive-compulsive and Avoidant personality disorders (Cluster C) have been described frequently (Braun et al., 1994). The relationship of Borderline Personality Disorder to bulimia nervosa has been a source of considerable debate (Wonderlich, 1995) with reported rates ranging from 2% to 47%, these rates apparently influenced by subject and measurement variability across studies, thus BPD probably occurs no more frequently than other PD’s in bulimia nervosa.
2.2.4 Course and prognosis There have been few studies with a lengthy follow-up period of the course and outcome of bulimia nervosa in the community. Many people with bulimia nervosa are not receiving any form of help (Hsu, 1995). Of these, the majority will suffer chronicity or a relapsing course, maintained by over- valued belief in the importance of appearance and thinness in particular (Fairburn et al., 2000). With the most effective treatments about 50% of people with bulimia nervosa can be expected to be asymptomatic two to ten years after assessment. Twenty percent are likely to continue with the full form of bulimia nervosa whilst the remainder (30%) have a course of illness characterised either by remissions or relapses or persistent but subdiagnostic bulimia (Hsu, 1995). One ten year follow-up study of 50 people with bulimia nervosa found that 52% had fully recovered and only 9% continued to experience symptoms of bulimia nervosa (Collings & King, 1994). A larger study of 222 followed-up for a mean of 11½ years revealed that 11% still met criteria for bulimia nervosa whereas 70% were in full or partial remission (Keel et al., 1999).
There are few consistent predictors of longer-term outcome, though a number of studies have shown strong associations between weight fluctuation (which commonly occurs in bulimia nervosa) and negative health outcomes. Patients with the uni-impulsive form of the illness without additional control difficulties are also thought to do better. The mortality rate associated with bulimia is uncertain but may be higher than in the matched general population (Hsu, 1995).
2.2.5 Bulimia nervosa in children and adolescents The full syndrome of bulimia nervosa is rarely seen in young people under the age of 14 (Bryant-Waugh & Lask, 1995). Indeed in these authors’ uniquely specialised clinic fewer than 5% of children under the age of 14 presenting with eating disorders were diagnosed with bulimia nervosa. Where these presentations occurred the clinical features were the same as found in the older age group. Young people with this disorder were also, in common with
22 CONFIDENTIAL Second draft for consultation September 2003
adults with the same diagnosis, depressed and suffered poor self image (Bryant-Waugh & Lask, 1995). There is no substantial literature on treatment or outcome in this age group. Adolescents with bulimia nervosa may sometimes be considered to be suffering incipient Personality Disorder, though caution should be exercised in diagnosing personality disorder before development is completed. In these cases it can be difficult to judge whether the eating disorder is contributing to abnormal personality development or conversely, if the personality difficulties have acted as a risk factor for the development of bulimia nervosa.
Since bulimia nervosa is very rarely seen in children and uncommonly in younger adolescents, and there has been no research on the treatment of adolescents with bulimia nervosa, in line with much current clinical practice and taking into account the above issues the GDG took the view that, subject to adaptation to their age, circumstances and level of development, adolescent patients with bulimia nervosa should receive the same type of treatment as adults with the disorder along with appropriate family involvement. 2.3 Atypical eating disorders (eating disorders not otherwise specified; EDNOS) including binge eating disorder
2.3.1 Symptoms, presentation & pattern of illness A number of people suffer with eating disorders which closely resemble anorexia nervosa and bulimia nervosa, but which are considered atypical, as they do not meet the precise diagnostic criteria for these conditions (Fairburn & Harrison, 2003, Turner & Bryant-Waugh, 2003 and Ricca et al., 2001). In Europe, these are often termed “atypical eating disorders” (Fairburn & Harrison, 2003), the equivalent American term being “eating disorders not otherwise specified” (American Psychiatric Association, 1994). For example, the patient’s weight might be just above the diagnostic threshold for anorexia nervosa or she might still be menstruating. Binge eating and purging may occur less frequently than specified for a diagnosis of bulimia nervosa. Over concern with weight and shape is generally present in these disorders, although in some the primary focus is on maintaining strict control over eating. Although the diagnostic criteria may not be met, many atypical eating disorders are as severe and long lasting as anorexia nervosa and bulimia nervosa. Patients’ treatment needs and prognosis may be virtually identical.
Binge eating disorder (BED) is a recently described condition, first defined as a research category in the 4th Revision of the American Psychiatric Association’s Diagnostic and Statistical Manual (DSM-IV; APA, 1994), though there is some overlap with the ICD 10 category Overeating associated with other psychiatric disturbance (F50.4) (WHO, 1992). In BED, individuals engage in uncontrollable episodes of binge eating but do not use
23 CONFIDENTIAL Second draft for consultation September 2003 compensatory purging behaviours. These binge eating episodes are associated with three or more of the following: • Eating much more rapidly than normal
• Eating until feeling uncomfortably full
• Eating large amounts of food when not physically hungry
• Eating alone through embarrassment at the amount one is eating
• Feeling disgust or extreme guilt after overeating.
Marked distress regarding binge eating is present and social avoidance is common.
2.3.2 Diagnosis Atypical eating disorders are conditions of clinical severity that do not conform to the diagnostic criteria for anorexia nervosa or bulimia nervosa. An example would be someone with extreme dietary restraint, who exercised excessively to control weight, which was maintained in the low normal range. This condition would also include those with the features of anorexia nervosa at low weight who are still menstruating. Many people with atypical eating disorders have suffered with anorexia nervosa or bulimia nervosa in the past. In comparison with anorexia nervosa and bulimia nervosa, far less is known about Binge Eating Disorder (Fairburn & Harrison, 2003). Apart from binge eating, its systematic profile overlaps little with the other eating disorders. Defining binge eating can be problematic and there may be a discrepancy between the subjective experience and clinical assessment of a binge. The onset of Binge Eating Disorder is usually in the teenage years or early twenties, but people tend to present later, typically in their 30s or 40s, when they have become overweight or obese. The sex ratio is more even and binge eating generally occurs against a background of a tendency to overeat rather than of dietary restraint. Many people with binge eating disorder are obese. By definition self-induced vomiting and laxative misuse are not present or only occasionally present. Depressive features and dissatisfaction with shape is common, though over-evaluation of the importance of weight and shape is less marked than in bulimia nervosa.
As with the other eating disorders, the diagnosis of Binge Eating Disorder is made on the basis of the history with physical investigations being used to assess any physical consequences. Observation and assessment in hospital is rarely indicated, though where available as part of a specialist day programme, assessment of meal preparation and eating may be a useful adjunct to treatment planning.
24 CONFIDENTIAL Second draft for consultation September 2003
2.3.3 Impairment and disability Where atypical eating disorders are similar to the full syndromes of anorexia and bulimia nervosa, the physical dangers and psychosocial impairments closely resemble those of the diagnostic conditions. Many of those with BED will suffer similar physical complications of binge eating to those with bulimia nervosa, though purging carries greater physical risk than binge eating. Those who are obese are at risk of the psychological and physical disabilities associated with this condition, namely low self-esteem, diabetes, heart disease, hypertension and stroke.
2.3.4 Course and prognosis The prognosis for those disorders that resemble anorexia and bulimia nervosa depends on the severity of the associated physical and psychological features. Those with binge eating disorder typically give long histories of proneness to binge eating but these may alternate with extended periods free from binge eating. The spontaneous remission rate may be high (Fairburn et al., 2000). Short-term response to treatment appears better than for anorexia nervosa and bulimia nervosa but there are as yet no studies of long-term course or outcome. There is a certain amount of movement from one diagnosis to another, thus those with atypical eating disorders may go on to develop bulimia nervosa, or more rarely anorexia nervosa.
2.3.5 Atypical eating disorders in children and adolescents Atypical eating disorders are relatively commonly diagnosed in childhood, in part because of the difficulty in strictly applying existing diagnostic criteria for anorexia nervosa and bulimia nervosa, and in part because of an ongoing lack of clarity about the classification of eating disturbances in this age group. Children and younger adolescents may present with a range of other types of clinical eating disturbance, which may be different in terms of psychopathology to anorexia nervosa/ bulimia nervosa presentations (Cooper et al., 2002). They frequently develop in those who have suffered feeding disorders of childhood and sometimes the distinction (particularly in middle childhood) can be a difficult one. Some have suggested that atypical eating disorders probably occur more commonly (Bryant-Waugh, 2000) and some have suggested modification to the ICD10 criteria for the diagnosis of eating disorders in children is required. BED has not been systematically investigated in this age group. 2.4 Incidence and prevalence of eating disorders The incidence of anorexia nervosa in the general population has been calculated from 12 cumulative studies at 19 per 100,000 per year in females and 2 per 100,000 per year in males (Pawluck & Gorey, 1998). The highest rates in this study were in female teenagers aged 13 to 19 years where there were 50.8 cases per 100,000 per year. A large Swedish cohort of 16-year-old schoolchildren who were assessed in a two stage screening process gave a prevalence for anorexia nervosa of 7 per 1,000 girls and 1 per 1,000 boys
25 CONFIDENTIAL Second draft for consultation September 2003
(Rastam et al., 1989). Anorexia nervosa has long been considered a culture bound syndrome; until recently the condition was thought to be extremely rare in developing countries and black populations Although there is growing evidence of the existence of eating disorders in a range of cultures, little is known of the incidence or prevalence in Asia, South America or Africa (Treasure & Schmidt, 2002), owing in part to differential rates of awareness and detection.
In community based studies, the prevalence of bulimia nervosa has been estimated between 0.5% and 1.0% in young women with an even social class distribution (Hay & Bacaltchuk, 2002). About 90% of people diagnosed with bulimia nervosa are female. The numbers of presentations of people with bulimia nervosa in developed countries increased steadily during the 1980’s and 1990’s, in addition, community surveys suggest a true increase in incidence during this period (Hall & Hay, 1991). The prevalence of eating disorders appears to be lower in developing countries and in rural areas. In Britain, young Muslim Asian women may be at particularly high risk of developing bulimia nervosa (Mumford & Whitehouse, 1988).
Less is known about the prevalence of binge eating than of bulimia nervosa (Fairburn, Hay & Welch, 1993). This is because two stage surveys designed to detect cases of bulimia nervosa have not generated figures for the prevalence of binge eating and secondly because of variability in interpretation of the term “binge eating”. Often self-report questionnaires are employed to elicit this behaviour with the possibility of discrepancy between lay and clinical uses of the term. Beglin and Fairburn (1992) have shown that there is relatively poor agreement between self-reported binge eating and the eliciting of this behaviour by the Eating Disorder Examination administered by interview. Probably because of these problems of definition and assessment Fairburn and Beglin (1990) found great variability in the figures available on the prevalence of binge eating. A review of 16 studies reported a range of 7% to 79% for current “binge eating” and 11 studies revealed a range of 5% to 39% reporting at least weekly binge eating. Within the normal weight sub group (i.e., those with a BMI below 25 kg/m2) only 1.7% of a subsequent study (Fairburn, Beglin & Davies, 1992) reported weekly binge eating. Thus, binge eating does not appear to be a common behaviour even among young women, though further research is necessary to estimate its prevalence with confidence.
Despite some shortcomings in epidemiological measurement in the area of binge eating, two large population based studies have estimated the prevalence of binge eating disorder using semi-structured interviews. Hay (1998) gives a prevalence of 1% for women, while Kinzl et al. (1999) estimate this at 3.3% for women and 8.5% for those with obesity. A further epidemiological study by Fairburn, Beglin and Davies (1992) sampled 285 women aged 16 to 35 years at random from General Practice case registers. In
26 CONFIDENTIAL Second draft for consultation September 2003 total, 243 agreed to be interviewed using the EDE. Twenty one (8.6%) reported at least one binge over the previous month whilst 10 reported at least an average of one weekly binge and 4 (1.7%) at least eight episodes in four weeks, a rate sufficient to make a diagnosis of bulimia nervosa or binge eating disorder. 2.5 The aetiology of eating disorders The aetiology of eating disorders in common with most other psychiatric disorders is generally considered to be multifactorial, no single aetiological factor in isolation can account for the development of the disorder in an individual, nor can it be seen to account for the variation among individuals (Cooper, 1995). Whether or not a person develops an eating disorder will depend on their individual vulnerability, consequent on the presence of biological or other predisposing factors, their exposure to particular provoking risk factors and on the operation of protective factors. Following the establishment of the disorder a further combination of risk and protective factors may act to maintain the condition or determine whether an individual recovers.
Much of the research in this area suffers from methodological weakness. Firstly much of it has been based on clinical samples often attending specialist clinics, which may result in selection biases. Few studies have included the appropriate control groups required to judge whether any putative aetiological factor is specific to eating disorders or might play a role in the development of psychiatric disorder in general. Secondly, a great deal of the published literature concerns cross-sectional or retrospective research. Where the onset of disorder is insidious, it is not always clear whether such factors are causes or consequences of the disorder. This is particularly true of family or life event research where the independence of any event may not be clear. Finally few studies have included the person with an eating disorder’s own perspective. One recent study (Tozzi et al., 2003), suggested that those with anorexia nervosa perceived dysfunctional families, dieting behaviour and stressful life events as the main causes of their condition.
A recent meta-analysis of prospective and experimental studies has systematically reviewed the evidence for aetiological and maintaining factors (Stice, 2002)
2.5.1 Genetic factors The majority of family studies have shown that eating disorders run in families. In a large case-control family study (Strober et al., 2000), female relatives of those with anorexia nervosa were 11.4 times as likely to suffer the disorder than relatives of control subjects, while female relatives of those with bulimia nervosa were 3.7 times as likely to suffer with bulimia. Some family studies have also reported familial aggregation of milder sub-diagnostic
27 CONFIDENTIAL Second draft for consultation September 2003 eating disorders or related concerns (Strober, 2000). The prevalence of full and partial bulimia nervosa has been shown to be more common in female relatives of those with anorexia nervosa than in relatives of control subjects and vice versa. These findings suggest that eating disorders form a spectrum of clinical severity in which there is a continuum of familial liability. The aggregation of full and subclinical eating disorders suggests that genetic factors are likely to be involved in causation. A twin study of anorexia nervosa has estimated the heritability to be 58% (95% Confidence Interval 33% to 77%) with the remaining variance explained by non-shared environment (Wade et al., 2000). Genetic factors are also purported to play a role in the aetiology of bulimia nervosa (Kendler et al., 1991). 54% to 83% of the variance in liability has been thought to be due to common genetic factors (Bulik et al., 2000), but again the confidence intervals around the estimation of heritability are broad, thus the relative contribution of genetic to other factors is unclear. The contribution of twin studies to the estimates of heritability of eating disorders have been criticised (e.g., Fairburn et al., 1999) on the grounds of ascertainment bias, small sample sizes or violation of the equal environment assumption which assumes that both identical and non-identical twins are equally exposed to aetiological environmental factors.
2.5.2 Physical risk factors A history of premorbid obesity has been documented in series of those with both anorexia nervosa (7 to 20%) and bulimia nervosa (18% to 40%) (Cooper, 1995) as has constitutional leanness/thinness. There is prospective evidence that this experience leads to a propensity to an increase in body dissatisfaction and likelihood of dieting behaviour (Stice, 2002) as well as greater self- evaluation in terms of weight. However there is little prospective evidence that high body mass leads to an increase in eating pathology.
Early feeding difficulties in infancy or early childhood are often documented but it is unclear whether this constitutes a physical risk factor or is related to the mother child relationship.
Although a range of neuro-endocrine and metabolic disturbances occurs in those with eating disorders, the evidence suggests that these disturbances are secondary rather than primary to the disorder.
Early menarche has long been considered a risk factor for eating disorder through a putative relationship with adiposity and body dissatisfaction.
The dietary restraint model suggests that calorie restriction increases the risk for binge eating and bulimia nervosa. Although dieting appears to increase negative affect and may contribute to eating difficulties, dieting has a small effect size in the contribution to the development of eating pathology (Stice, 2002).
28 CONFIDENTIAL Second draft for consultation September 2003
2.5.3 Adverse life events and difficulties Severe life stresses have been implicated in the aetiology of both anorexia nervosa and bulimia nervosa, with approximately 70% of cases being triggered by severe life events or difficulties. These stresses most commonly occur in the area of close relationships with family or friends (Schmidt et al., 1997, Welch et al., 1997). One controlled study of anorexia nervosa in adolescents (North et al., 1997) suggested that a severe independent negative life event had occurred in the year before onset of approximate one quarter of an adolescent series and this was associated with a good prognosis. Particular attention has focussed on the experience of childhood sexual abuse as a potential predisposing factor. There is little persuasive evidence that either sexual abuse or other stressful life events are specific predisposing factors for eating disorders rather than psychiatric disorder per se. Childhood sexual abuse did not emerge as a significant predictor of the onset of binge eating in the only prospective study to date (Vogelantz-Holm et al., 2000)
2.5.4 Family factors Mounting evidence from family studies suggests increased rates of Affective Disorder among first and second degree relatives of people with both anorexia nervosa and bulimia nervosa. This is three times greater than for relatives of normal control participants (Cooper, 1995). Several family studies have revealed higher rates of alcohol abuse among first degree relatives of those with bulimia nervosa but again, the specificity to eating disorder as opposed to psychiatric disorder in general is unclear. A family history of substance abuse may however, be a specific risk factor for bulimia nervosa (Cooper, 1995).
A number of studies have looked at family environment and functioning. There are difficulties extrapolating from disturbance within the family after the disorder has developed and assuming that it predated the onset of disorder. Prospective studies have to date failed to provide support for the causative role of dysfunctional family systems (Stice, 2002). Familial faddy eating and undue concern about the importance of weight and shape have been described in family members of those with eating disorders and may contribute to weight and shape concern in vulnerable adolescents (Gowers & Shore, 2001). There is also growing evidence of disturbance of eating in the children of mothers with eating disorders (Stein, 1995) but it is as yet unclear whether these children will be at risk of anorexia and bulimia nervosa in their own adolescence. One controlled study demonstrated a style of “high concern” parenting in a number of adolescent cases of anorexia nervosa, which arose in infancy and long-predated the onset of eating difficulties. In a quarter of cases, this followed an earlier severe obstetric tragedy prior to the subject’s birth. These rates were significantly higher than in a matched control group (Shoebridge & Gowers, 2000) and challenge the notion that observed parenting styles are solely a response to the development of an eating disorder in adolescents.
29 CONFIDENTIAL Second draft for consultation September 2003
2.5.5 Socio-cultural factors A number of socio-cultural theories have been put forward to explain the aetiology of eating disorders. Such theories include the meaning of weight and shape for women in different cultures and the impact of advertising and other media. It is argued that societal pressure to be thin fosters an internalisation of a thin ideal and body dissatisfaction, which in turn leads to dieting behaviour and places the person at risk for eating pathology (Striegel- Moore et al., 1986). This perceived pressure does appear to predict dieting and eating pathology (Stice, 2002). Furthermore, in prospective studies, perceived pressure to be thin predicted the onset of binge eating, bulimic symptoms and increases in eating pathology (Stice, 2002).
2.5.6 Perfectionism This personality trait has long been considered a risk factor for eating pathology as it may promote the relentless pursuit of the ‘thin ideal’. Fairburn (1997) has implicated perfectionism as a maintenance factor for bulimic pathology as it encourages rigid dieting thought to underlie the binge-purge cycle. Meta-analysis of prospective studies provides support for the notion of perfectionism as a risk factor for bulimic pathology and a maintenance factor for more general eating pathology (Stice, 2002).
2.5.7 Impulsivity General problems in the area of impulse control have been proposed as risk factors for eating pathology as they may render the individual vulnerable to episodes of uncontrollable binge eating (Hawkins & Clement, 1984). The empirical support for this association is weak (Stice, 2002), however some indirect support is offered by the finding of an association between substance use and bulimic pathology (Stice & Agras, 1998)
2.6 Use of health service resources Eating disorders can be chronic, recurrent mental disorders with important psychiatric co-morbidities (depression, generalized anxiety disorder, alcoholism, phobias, panic disorder and posttraumatic stress disorder) and physical complications such as cardio-vascular and renal problems, gastrointestinal disturbance, fluid and electrolyte abnormalities, menstrual and fertility problems, osteoporosis and osteopenia, dental and dermatological abnormalities (Fairburn & Brownell, 2001). As such they can place considerable burdens on the health care system. Eating disorders are at 15th place among the top 20 causes of disability in Australian women measured in years lost due to disability (Vos et al., 2001). Similar rates probably apply in the UK.
In an Australian burden of mental diseases study (Vos et al., 2001), the estimated average duration for anorexia and bulimia nervosa was eight and five years, respectively. These assumptions were based on follow-up studies
30 CONFIDENTIAL Second draft for consultation September 2003 reporting 24% persistence of anorexia nervosa after 10-15 years and 20% persistence of bulimia nervosa after 5-10 years (Strober et al., 1997; Keel & Mitchell, 1997). Anorexia nervosa has the highest mortality rate for any psychiatric condition from the effects of starvation or by committing suicide. At 10-year follow-up the standardised mortality ratio was 6%, at 20-year follow-up the mortality rate was estimated to be 13-20% (Howlett et al., 1995). The annualised mortality rate for individuals with anorexia nervosa was found to be 0.56% in a meta-analysis (Sullivan, 1995) and it was not elevated in bulimia (Nielsen et al., 1998). Very little is known about the mortality associated with BED and EDNOS, although the obesity common in BED is expected to increase the mortality risk (Crow & Peterson, 2003).
Physical complications and mortality in these illnesses are well recognised, but other social, occupational and economic costs and the negative impact on quality of life (Keilen et al., 1994) have received far less attention. The chronic nature of eating disorders and the numerous co-morbidities and complications suggest that people with eating disorders need multi- dimensional treatment and are high consumers of medical and social care. However, a comprehensive study of health or social services use of these individuals has not been carried out either in the UK or internationally (Garvin, 2001). Only a limited literature exists about the broader costs of eating disorders in the healthcare system and this research evidence indicates significant economic burden posed by eating disorders.
2.6.1 Primary care Studies from the USA and continental Europe suggest that only a fraction of people with eating disorders receive specialised treatment for their eating disorders. For example, in a large sample of female primary care and obstetric gynecology attendees in the USA fewer than 1 in 10 cases with bulimia nervosa or BED were recognized by the physician (Johnson et al., 2001). In the Netherlands, on average, only 40% of the community cases of anorexia nervosa are detected by general practitioners and 79% of these patients are referred on for mental health care (Hoek, 2003). In the case of bulimia nervosa, only a small subgroup, an average 11% of the community cases, are detected in primary care and 51% of these cases are referred to specialist care. A general practitioner with a list of 2000 people could expect to have one or two patients with anorexic nervosa and 18 patients with bulimia (Hoek, 1991). Importantly, the poor detection rates for eating disorders do not suggest that these patients avoid their general practitioners. Indeed, over the 5 years prior to the diagnosis of the eating disorder, these individuals consult their general practitioners significantly more frequently than do people without an eating disorder (Ogg et al., 1997). In these earlier consultations, patients typically present with psychological, gastro-intestinal or gynaecological complaints prompted by symptoms of the eating disorder or its complications. These findings suggest that there are high levels of hidden eating disorder morbidity at primary care level leading to a considerable underestimate of
31 CONFIDENTIAL Second draft for consultation September 2003
costs, and greatly emphasise the importance of primary care in the detection and early treatment of eating disorders.
In the UK, an epidemiological study using the general practice research database found similar figures for referrals, 80% of cases of anorexia nervosa and 60% of cases of bulimia nervosa were referred on to secondary care, the majority to psychiatrists. The study also revealed that general practitioners tend to prescribe laxatives or diuretics for 27 of 100 people with the diagnosis of bulimia nervosa and psychotropic medication in 45% of the eating disorders cases (Turnbull et al., 1996) It should also be noted that many GPs may fail to consider an eating disorder as possible diagnosis in children presenting with typical eating disorder features (Bryant-Waugh et al., 1992). The only calculation of the total cost of eating disorders in primary care was done based on the 3rd National Survey of Morbidity in General Practice which found that 1/1000 females and 0.6/1000 males consulted their general practitioner for anorexia nervosa (no data for bulimia nervosa, EDNOS or BED were available) in one year (Office of Health Economics, 1994). Applying these rates to the 1990 UK populations suggest that approximately 47,000 individuals consult their GP with anorexia nervosa each year, which yields an estimated cost of £580,000 for anorexia nervosa consultations in primary care in 1991 prices (Office of Health Economics, 1994).
2.6.2 Secondary and tertiary care A recent North American study analysing a national insurance claim database found that specialist eating disorder treatment provision is rare and eating disorders usually remain undetected or undertreated especially in the case of men (Striegel-Moore, 2000). A UK study of patients with anorexia nervosa revealed that under-diagnosis is common. Some patients may spend months or years in non-specialist units posing enormous extra costs for the NHS (Howlett, 1995). The NICAPS study (O’Herlihy et al., 2001) has shown that more beds were occupied by young people with EDs than any other diagnostic group, with the same number of patients with eating disorder in specialist eating disorders services being found as in non-specialist psychiatric units. In all, there was a total of 22.2% of the inpatient population with a diagnosis of an eating disorder.
In the USA as in the UK, outpatient treatment is the norm with low hospitalisation rates for eating disorders. Striegel-Moore (2000) reviewed costs for the disorders in the USA. The average numbers of outpatient episodes are 17, 15.6 and 13.7 for anorexia nervosa, bulimia nervosa and EDNOS, respectively. Only 21.5% of women with anorexia nervosa are hospitalised per year with an average length of stay in the US of 26 days, and very much lower hospitalisation rates were found for bulimia nervosa and EDNOS. Mean annual treatment (outpatient and inpatient) costs for female patients were $6,045 for anorexia nervosa, $2,962 for bulimia nervosa and $3,207 for EDNOS. These costs of treating eating disorders were compared with treatment costs of schizophrenia and obsessive-compulsive disorder
32 CONFIDENTIAL Second draft for consultation September 2003
(OCD). The mean treatment cost for anorexia nervosa was significantly higher than the mean cost for schizophrenia, but mean treatment cost for bulimia nervosa was significantly lower than for schizophrenia. Treatment costs for any eating disorder were much greater than that of OCD These cost estimates are likely to represent an underestimate of the true costs of these illnesses since costs were calculated based on diagnostic codes (Crow & Peterson, 2003).
In the UK in 1985 an estimated 25,748 bed days were allocated to female anorexia nervosa inpatient treatment (Hospital Inpatient Enquiry) with the average length of 21.5 days per episode. Only a fraction of the medication costs were estimated in the same study. There are no more recent UK estimates about inpatient treatment, other medication, outpatient treatment and psychotherapy costs, and no data are available for bulimia nervosa, EDNOS or BED. When looking at the organisation of eating disorders care in the UK, it is important to mention the significant role of private providers even though they are acknowledged as being a more expensive way to contract with providers (Brown, 1997).
The most recent estimate of the health care costs of eating disorders comes from Germany. Krauth et al. (2002) calculated that the health care costs of anorexia nervosa for the year 1998 amounts to 64.9 million euros and that of bulimia nervosa to 9.8 million euros without the costs of primary care, outpatient care (including psychotherapy) and pharmaceuticals. The estimated hospitalisation cost of 12,800 euros per anorexia patient was found to be much higher than the hospitalisation cost of 3,600 euros for an average patient with bulimia nervosa (Krauth et al., 2002). However, it is worth mentioning that in Germany the threshold for inpatient admission for anorexia nervosa is probably much lower than in the NHS in the UK if not in the private sector.
The cost of managing physical and dental complications (such as infertility, osteoporosis and reconstructive dentistry), in secondary care has not been quantified but may be considerable.
2.6.3 Non-health care burden Eating disorders have a substantial impact on social functioning, including occupational and educational impairment. These disorders are the most common at an age when people are in secondary school, in higher education or at the beginning of their working careers. Eating disorders often result in lost productivity due to the inability of people to work or premature death. The lost productivity costs of anorexia nervosa were estimated to be 130.5 million euros (inability to work: 8 million euros, death: 122.5 million euros) for the 82 million German population in the year 1998. The same figure for the indirect economic burden of bulimia nervosa was 113.9 million euros
33 CONFIDENTIAL Second draft for consultation September 2003
(inability to work: 1.7 million euros, death: 112.2 million euros) (Krauth et al., 2002).
The average duration of an episode of anorexia nervosa is 6 years (Herzog et al., 1997) and the family are usually the main carers for most of the duration of the illness. A recent UK pilot study found that carers of anorexic patients reported similar experiences in terms of the difficulties experienced to those of carers of adults with psychosis, but had significantly higher level of psychological distress (Treasure et al., 2001). The burden of care giving has never been examined in economic terms.
These estimates suggest that the broader social costs of eating disorders may be even more substantial than their health care costs, and that early diagnosis, prevention of chronicity and appropriate and cost-effective treatments may greatly reduce the broad economic burden of eating disorders. 2.7 The treatment and management of eating disorders in the NHS
Until the last few decades of the 20th century, in the UK, general physicians generally managed anorexia nervosa. Treatment consisted of bed rest, often in a side room of a general medical ward. A few psychiatrists started to take an interest in the condition from the middle of the 20th century. Dally and Sargant used physical methods of treatment such as insulin and chlorpromazine (Dally, 1969). Then two academic psychiatrists, Professor Arthur Crisp and Professor Gerald Russell, took an interest in the condition initially with a particular interest into the endocrine and metabolic aspects. They developed specialised units, which were replicated across the UK and the world and trained most of the current generation of psychiatrists working in eating disorder services Skilled, supportive nursing was the cornerstone of such units but over time more specific forms of psychotherapy were developed. At the Maudsley Hospital, for example there was a particular interest in family therapy.
The description of bulimia nervosa by Gerald Russell in 1979 was a landmark event. An explosion of research followed this into both psychotherapeutic and physical (antidepressants and appetite suppressants) treatments. The most widely researched treatment has been cognitive behaviour therapy developed by Chris Fairburn.
At the present time, there are wide variations in the provision of eating disorder services and models of service delivery throughout the UK. Services range from ‘generic’, in which outpatient therapies are provided by community mental health teams, backed up by psychiatric admission, to variable models of specialist eating disorder service. These latter vary widely in staffing, service configuration and therapeutic interventions offered. At the
34 CONFIDENTIAL Second draft for consultation September 2003
more modest end of the spectrum, an outpatient only service may offer therapeutic input, but little in the way of medical assessment or management, responsibility for which remains with primary care. More fully comprehensive models are organised around intensive day-patient provision, an inpatient unit, or both. Many of the better-funded resources function as tertiary services, serving a large (on occasions, national) catchment area.
Children and younger adolescents (up to the age of 16) are generally treated in generic child and adolescent mental health services (CAMHS) supported as required by paediatric services. Where inpatient treatment is thought necessary, children and adolescents are admitted either to paediatric wards often with CAMHS liaison, to ‘general-purpose’ children’s or adolescent units, or to private-sector specialist eating disorder services, funded by the NHS.
Where young people are admitted to paediatric medical wards, appropriate communication with CAMHS staff is crucial. The care plan should be agreed between the medical and mental health services and appropriately co- ordinated. There are very few specialist eating disorder inpatient services for this age group in the NHS and none with a comprehensive range of out- day and inpatient provision. Many services attempt to avoid inflexible age boundaries, preferring to take into account the young person’s developmental level and whether they are in full time education or living at home with parents when deciding the appropriate service for them.
Ensuring appropriate communication and handover across transitional boundaries is a particular challenge for those treating children and adolescents. Some services offer combined adolescent / adult treatment, which offer experience in the management of eating disorders, but need to ensure developmentally appropriate therapeutic intervention and social support. In many parts of the country, the potential advantage of specialist inpatient treatment may be offset by the disadvantages of treatment a long way from home, with problems posed for therapeutic involvement of the family, maintaining educational and social contacts and most importantly, ensuring continuity of care post-discharge.
2.7.1 Pharmacological treatment Pharmacotherapy has not been the treatment of first choice for eating disorders, but it has been used as an adjunct to psychological therapies or to treat physical or co-morbid psychological problems. In anorexia nervosa, medication has to date been disappointing in influencing the core symptoms of the disorder, promoting weight gain or reducing associated mood disturbance, (Treasure & Schmidt, 2001). Where modest improvements have been reported, consideration of unwanted effects (e.g., in prolonging the QT cardiac interval), have led researchers to conclude that drugs confer little advantage when added to standard treatment, (e.g., Vandereycken, 1984). In
35 CONFIDENTIAL Second draft for consultation September 2003
bulimia nervosa and binge eating disorder, whilst drugs are not the first option, there is some evidence that antidepressants, particularly selective serotonin reuptake inhibitors (SSRI’s) contribute to the cessation of binge eating and purging. As endogenous opiates play a role in the hypothalamic regulation of hunger and satiety, a few trials have been conducted with opiate antagonists.
2.7.2 Psychological interventions Choosing the most effective treatment approach for someone with an eating disorder should take account of a number of variables, including physical and psychological risk, motivation, social support, comorbidity, and age. Often and particularly in anorexia nervosa, treatment planning will require co- ordinated, multidisciplinary, physical, psychological and service interventions. Psychological interventions are however considered to be crucial in addressing the core attitudes that underlie these disorders and in influencing the longer-term outcomes. The nature of the psychological therapies chosen will be influenced by patient preference, their motivation, the nature of associated psychological features and their age or stage of development. Some will prefer a non-verbal projective therapy, using art, drama or music. Younger patients in particular or those who are dependent on relatives or carers are seen as often requiring family or systemic therapy. Of the individual therapies, cognitive behaviour therapy (CBT) is the best researched, though the evidence base in bulimia nervosa and binge eating disorder (BED) is far larger than in anorexia nervosa. The evidence base for other psychological therapies such as supportive psychotherapy and interpersonal psychotherapy (IPT) is growing, while more specialised therapies such as Cognitive Analytic Therapy (CAT) and Dialectical Behaviour Therapy (DBT) have been considered in complex cases and for DBT, those complicated by repeated self-harm (Treasure & Schmidt, 2003).
2.7.3 Service-level interventions Decisions about the right treatment setting in which to manage a person with an eating disorder currently depend on the nature of the disorder, the level of risk, physical and psychological complications and patient preference. Within the NHS currently, the availability of different models of service is a major consideration with patients and referrers sometimes having to weigh up the advantages of a locally accessible service with those of a more comprehensive service at some geographical distance. In general, those with bulimia nervosa and binge eating disorder will be treated on an outpatient basis either in primary care or secondary services. Those with anorexia nervosa will generally be treated in secondary care, the choice of in-, out-, or day patient provision depending on the above considerations. Most will have a trial of outpatient intervention first. A stepped-care model in which patients move up from secondary to tertiary (specialist) care subject to locally agreed protocols makes clinical sense, but there is at present little evidence to guide decisions about service setting. There is also considerable debate over the
36 CONFIDENTIAL Second draft for consultation September 2003
requirement for some or all of people with eating disorders to be treated in specialist services. In general the current consensus would seem to favour the treatment of the large majority, if not all patients with anorexia within specialist services, whereas specialist services may only be needed for patients with severe forms of bulimia nervosa. However, irrespective of the service setting (specialist or generalist) the key aim of this guideline will be to ensure that as far the evidence allows effective treatments should be provided. Whatever the service, the setting should be age-appropriate, with the social, and educational needs of children and adolescents requiring particular attention in order to avoid the development of secondary handicaps Inpatient management will require comprehensive physical and psychological management; currently a number of models exist from medical admission with psychiatric/psychological liaison to psychiatric admission with medical supervision. Children, in particular are often admitted to general medical (paediatric) services.
2.7.4 Primary–secondary care interface Many patients with eating disorders will require treatment for a considerable time, often for a number of years. During this time, they may require treatment in primary care, secondary and on occasions, tertiary care services. The boundary between primary and secondary care is one of a number of interfaces that need to be managed effectively. Others include the interface between medical and psychiatric services, between inpatient and outpatient services, and for adolescents, between Child & Adolescent Mental Health Services (CAMHS) and adult services. It is particularly important for effective management that communication is good and that areas of responsibility are clear. Sometimes, for example a patient may be receiving psychological therapy from a secondary care service, but responsibility for physical monitoring may remain with primary care. Clear treatment protocols will ensure smooth management across these interfaces.
2.7.5 Physical health care Patients with eating disorders always require assessment of their physical health and any associated risk as part of a comprehensive assessment. This should also be reviewed as treatment progresses. Physical complications can be classified as follows: • The complications of starvation (chiefly in anorexia nervosa)
• The consequences of purging behaviours (including dental complications)
• The complications of associated physical conditions (such as diabetes or pregnancy)
• The complications of excessive exercise
37 CONFIDENTIAL Second draft for consultation September 2003
• Physical aspects of stunted or incomplete development (in younger people); the main effects being on skeletal growth and the reproductive system
• Physical complications of dietary imbalance (e.g., a high fibre, low fat diet)
• Physical complications of re-feeding or restoration of normal diet.
Many of these may have effects on a number of systems within the body. In any treatment plan, it must be clear who is taking responsibility for physical assessment and how any risk identified is to be managed. This often involves effective communication between primary and secondary care services (see section 2.7.4).
2.7.6 The relationship of the evidence base for adults to that for children and adolescents. In the absence of a strong body of treatment research in the child and adolescent literature, it is tempting to extrapolate from adult findings, but the validity and limitations of doing so requires careful consideration. A key argument in favour of extrapolating from adult findings is that adolescence is a developmental stage that is not defined merely by age. It can be argued that many young adults with eating disorders are still in the throes of addressing the challenges of adolescence and indeed developmental difficulties have been thought to underlie the aetiology of anorexia nervosa in particular (e.g., Crisp, 1995). A second argument is that the essential features of anorexia nervosa and bulimia nervosa are consistent across the age spectrum – both in terms of characteristic behaviours (dieting, binge eating, purging), specific psychopathology (over-evaluation of the self in terms of weight and shape) and non-specific features (low self-esteem, perfectionism, poor interpersonal confidence). Thirdly some of the literature reports combined adolescent / adult case series without separate analysis. Finally some of the treatments that have been found to be effective in adult eating disorders (e.g., CBT and antidepressants) are effective in the treatment of adolescents with other conditions (particularly depression).
There are however a number of reservations about extrapolating from adult findings. In younger patients, the disorders less commonly fall neatly into the ICD10 or DSM-IV categories; that is to say, atypical forms are more commonly diagnosed. Some clinical eating disturbances seen in childhood and early adolescence currently classified as atypical eating disorders may not have much in common with the adult category of EDNOS. As treatments for EDNOS are poorly developed in adults, there are few findings in any case to draw on. The treatment aims, particularly in anorexia nervosa are often different in childhood and adolescence, because of the different physical issues involved i.e., where the onset is before growth and development are
38 CONFIDENTIAL Second draft for consultation September 2003
complete, treatment needs to address the completion of puberty and growth. Unlike the treatment of adults with anorexia nervosa where recovery usually involves returning to a pre-morbid healthy physical state, in younger patients it is more a question of coming to terms with a new physical state. In terms of weight targets this involves ‘constantly moving the goal posts’ as healthy weight is recalculated with the attainment of greater height. All this might indicate a need for a longer duration of treatment for younger cases, whilst a shorter duration of illness before treatment might argue for the opposite.
When considering the literature on pharmacotherapy, one should be aware of differences in the pharmacodynamics and pharmacokinetics in children. In general the latter means that children and adolescents require higher doses of drugs per kg body weight to attain similar blood levels and therapeutic effect, owing to children’s more rapid metabolism by the liver and clearance by the kidney.
Irrespective of any consideration of aetiological variables, parents will usually need to be involved in the management of younger patients. This is especially so if they are at risk and parental involvement is likely to reduce the risk. Both anorexia nervosa and bulimia nervosa require aspects of behavioural management and parents will need to be involved if management of these is to be effective; at a practical level, parents usually have a role in shopping for food, meal planning, and managing meal times. The involvement of siblings is generally regarded as beneficial, for the sibling if not for the patient, as this provides an opportunity for them to express fears or guilt and to dispel any false ideas about the nature of the condition, its likely causes and prognosis.
Additional attention will need to be given to the different social and educational needs of this age group in treatment, particularly when treated in hospital.
In addition, carers and parents of adolescents have identified a lack of and need for support, involvement and education about eating disorders for themselves (Kopec-Schrader et al., 1993; Haigh & Treasure, 2003). 2.8 Assessment Eating Disorders may present to the NHS in primary care or to a range of tertiary medical services including mental health, gastroenterology, endocrinology and reproductive medicine. Some patients will volunteer symptoms of eating pathology and disclose the characteristic behaviours, or a relative or other carer will report them (particularly in childhood and adolescence). Others will present with the symptoms of physical or psychological complications either because they are unaware that they are symptoms of an eating disorder, or because they do not wish to reveal it. A most important step in the identification of eating disorders therefore is to be open to the possibility of the disorder (see section 5.1).
39 CONFIDENTIAL Second draft for consultation September 2003
Assessment encompasses making a diagnosis and eliciting the necessary information to prepare a care plan, including the assessment of severity and risk. The details of assessment methods are provided in the primary care chapter (Sections 5.2.4.1, 5.2.4.2). Wherever assessment takes place however, some common principles apply: • Keeping an open mind
• Maintaining a supportive, non-judgmental approach
• Focusing on engagement (see section 2.9)
• Assessing risk (see sections 6.4.3, 7.5)
• The patient’s right to confidentiality (see section 2.11.2).
As eating disorders are commonly characterised by ambivalence, assessment should not be seen in isolation from treatment, as the patient’s first impressions of services may have a powerful impact on their willingness to consider referral to secondary care and subsequently accept treatment. Specialist services are increasingly adopting a motivational approach to assessment and induction into treatment (Treasure & Bauer, 2003). Throughout this guideline, reference is often made to the comprehensive assessment of physical, psychological and social needs. This should be taken as a general indication as to the content of a comprehensive assessment which will vary form person to person but will often include: • Current and past physical health and treatment
• Cognitive capacities
• Any present physical disabilities
• A historical and current assessment of family and interpersonal relationships
• Mental state and personality factors
• Social circumstances and supports
• Occupational and financial status
• Educational needs (for children and adolescents only).
40 CONFIDENTIAL Second draft for consultation September 2003
2.8.1 Clinical practice recommendations
2.8.1.1 Assessment of people with eating disorders should be comprehensive and include physical, psychological and social needs, and a comprehensive assessment of risk to self. (C)
2.8.1.2 The level of risk to the patient’s mental and physical health should be monitored as treatment progresses as it may increase – for example following weight gain or at times of transition in cases of anorexia nervosa. (C)
2.8.1.3 Healthcare professionals assessing children and adolescents with eating disorders should be alert to indicators of abuse (emotional, physical and sexual) and should remain so throughout treatment. (C)
2.8.1.4 The right to confidentiality of children and adolescents with eating disorders should be respected. (C)
2.8.1.5 Healthcare professionals working with children and adolescents with eating disorders should familiarise themselves with national guidelines and their employers’ policies in the area of confidentiality. (C)
2.9 Engagement, consent and the therapeutic alliance Many people presenting to NHS services are ambivalent about revealing their symptoms and behaviours and are uncertain about or indeed actively opposed to treatment. There are a number of reasons for this ambivalence. Firstly, the patient may believe that they will be criticised or treated unsympathetically. This belief may be based in guilt about engaging in vomiting or binge eating, or a belief that they will be expected to exert control over these maladaptive behaviours. Alternatively they may have heard about the negative treatment experiences of others (see section 2.10), or fear compulsory treatment. Anorexia nervosa in particular is sometimes perceived by the patient as a valued life-style choice or the only possible way for them to live. Many will fear the effects of treatment; someone with anorexia nervosa may believe that treatment will make them fat, whilst a patient with bulimia nervosa may fear that stopping vomiting will render them vulnerable to impulsive binge eating without recourse to compensatory purging.
A style of empathic engagement is helpful therefore in addressing patient anxieties, but probably also has an important bearing on treatment outcome. Although considerable attention may be necessary to address the physical complications of eating disorders, these interventions rarely address the core attitudes and behaviours underlying the disorders, which are generally addressed by psychological therapies. As in the treatment of any disorder, effective psychological therapy requires engagement in a therapeutic
41 CONFIDENTIAL Second draft for consultation September 2003
relationship and agreement about treatment aims. Whilst a forceful approach may result in a degree of weight gain in anorexia nervosa, clinicians are increasingly drawing attention to the importance of engagement and positive motivation if short-term gains are to be maintained in the long term, whatever the treatment setting (Ward et al., 1996).
In most cases for adults treatment is offered on the basis of fully informed consent. In rare cases, almost exclusively of anorexia nervosa, this is not possible. Anorexia nervosa strikes at one of the core aspects of being an autonomous individual, the capacity to care for oneself and maintain health and safety. The Department of Health has recognised that a patient with anorexia nervosa who is unable to appreciate his or her failing condition may not be able to comprehend and make use of relevant information and as a result may lack the capacity to make a valid treatment decision, (Department of Health, 2002a). This is recognised by society by the use, on occasions, of the Mental Health Act in this condition. It is good practice however, whether treatment is provided on a formal or informal basis to develop a therapeutic alliance, attempt to promote positive motivation and engage in informal treatment as soon as the physical and psychological state of the patient give confidence that informal treatment is likely to be effective.
Similar principles apply to the treatment of children and adolescents. The transition from childhood to adulthood can however cause confusion about rights and responsibilities. Potential additional difficulties arise from the need to inform parents and carers about risks, but maintain, as far as possible, the young person’s right to confidentiality (see section 2.11). The government specifically addresses issues relating to consent and confidentiality as it relates to this transitional phase. “Young people aged 16 and 17 are regarded as adults for the purposes of consent to treatment and are therefore entitled to the same duty of confidentiality as adults. Children under the age of 16 who have the capacity and understanding to make decisions about their own treatment are also entitled to make decisions about the use and disclosure of information they have provided in confidence (e.g., receiving treatment and counselling that they do not want their parent to know about). However where a competent child is refusing treatment for a life threatening condition, the duty of care would require confidentiality to be breached to the extent of informing those with parental responsibility for the child, who might then be able to provide the necessary consent to the treatment (Department of Health, 2002a). 2.10 The patient experience of eating disorders
For patients, the experience of an eating disorder can be complex, ambivalent, and contradictory. Complex, because a range of issues must be untangled and understood on the often long road to recovery; ambivalent, because of the tension between wanting to get better while fearing the implications of
42 CONFIDENTIAL Second draft for consultation September 2003
recovery; and contradictory because the disordered eating is at once the problem and the solution - a destructive lifestyle that nevertheless keeps the person in control, safe and protected, while offering a dependable and consistent presence in their life.
Nobody just wakes up one day with an eating disorder. These are conditions that develop over time, sometimes over years and often at a point when life changes bring fear and insecurity. Both anorexia nervosa and bulimia nervosa often start with a period of food restriction of some kind, which gradually becomes stricter and stricter. Such restrictions often are reported as generating a sense of euphoria at being in control of one’s weight, of feeling superior to others who might be struggling to manage theirs, and of increased confidence and enhanced appearance. Patients report that it is not unusual to feel moments of intense clarity, insight and of feeling more alive than ever. It is also not unusual to feel considerable physical energy, which may be experienced as though activity could be sustained for many hours without food or sleep; at least until the food restriction starts to trigger physical and mental side effects. All this may serve to reinforce a person’s negative behaviours around food.
At the same time, patients often sense that all of this is somehow not quite right. While recognising that their eating patterns are unhealthy in terms of quality or quantity of food, they may dread eating more for fear of gaining weight or being unable to stop eating once started. They may become increasingly secretive and devious around food and eating – skipping meals, lying about eating, hiding food, eating in secret, discovering ways to get rid of food once eaten, while denying or not acknowledging that there is a problem. Food increasingly comes to dominate thoughts, feelings and actions. Debating the options of what to eat becomes a constant daily struggle, an internal battle, a desperate self-involved fight where even sleep offers little escape. Not eating is somehow so much easier than eating, but for so many this often ends in chaos, binge eating and weight gain.
Self-hatred about appearance, feelings of being bloated and fat and frequent physical pain are common consequences of these behaviours. Starting to eat can worsen these feelings. Even those who don’t get caught in the starve/binge cycle frequently lose touch with friends and family and become increasingly alone and isolated because of the difficulty of discussing these feelings for fear that no one could possibly understand. The person may start to be caught between extreme control and the overwhelming fear of descending into complete chaos – knowing that at any time “giving in” to food is only a step away. If the control slips, the chaos takes over and shame and guilt flow in its wake.
Many of these feelings are common, whether the diagnosis is of anorexia, bulimia or binge eating. The person may be aware of a deep underlying despair, a sense of being inherently inferior, unworthy and undeserving of the
43 CONFIDENTIAL Second draft for consultation September 2003
good things in life, guilty for wanting to try to fulfil their needs. Yet despite constant attempts to deny the problem and the difficulties inherent in loving, needing and self-acceptance, there is, somewhere, an intense desire to be loved, needed and accepted as one is.
The person may reach the point of acknowledging the problem, and turning to someone to ask for help. This is a critical first step towards recovery although the way ahead may still be fraught with problems. A range of questions arise: “who is trustworthy – doctor, teacher, nurse, parent, friend, relative …? Who will really listen and take me seriously? Who will see me as a whole person and not just as an illness or a difficult problem? Who will understand that someone really can have these issues with food? What will happen if I ‘come clean’ and admit to my behaviours? What will a diagnosis mean in terms of the treatment I might be offered? How much control will I retain? And will admitting to a problem, blight the landscape of my future career, work prospects, education, relationships?”
For boys and men, there may be added complications since, despite changing attitudes and understanding, eating disorders are still considered primarily a female issue. This can make it even more difficult for men to seek help. For young people, there are often particular questions surrounding confidentiality – “If I speak to someone at school or my doctor, will they tell my parents?; with control of treatment - given my age, who will be ‘in charge’? And with the fear of being treated differently to one’s peer group – ‘you have a problem with food, therefore you cannot be trusted!’”. All of these issues, together with low confidence and self-esteem may convince someone that keeping the disorder hidden is the safest option, even if, deep down, they want to move forward. They may just not believe that they are strong enough to do so.
Being given a diagnosis can raise a new set of issues. Treatment in mainstream services can be a postcode lottery with long waiting lists. It can be difficult, if not impossible, to learn about the range of treatment options and services if the GP lacks experience of working with eating disorders. There may be a debate about how best to treat the individual, and treatment may be only available at some distance from family and friends. The time-lag between raising the issue, getting a diagnosis and accessing treatment can leave the person with an eating disorder feeling ‘let-down’ or rejected and compound feelings of unworthiness after having found the courage to speak out. Any delay also provides an opportunity for ‘ambivalence’ to creep back in again – the person may then ask “Do I really want this treatment anyway?”
Eating disorders present complex challenges and treatment tends to be long- term with the possibility of frequent relapses. Whatever the setting, finding the ‘right person’ to work with – someone with expert knowledge and particular qualities who will accept and understand the person as ‘an individual with a unique experience’ rather than as ‘a case of pathology’ -
44 CONFIDENTIAL Second draft for consultation September 2003
seems critical to treatment ‘success’. Trust is also vitally important when deciding to give up the principal means to cope with life. Getting help with food itself, through dietary counselling, can also be important particularly with bulimia. It can be very difficult for an individual to focus on psychological and emotional processes when thoughts are crowded with what one will/won’t, can/can’t eat, or when the body is in a state of starvation, poor blood sugar control, hormone/neurotransmitter imbalance etc. Establishing a structured approach to food and being given permission to ‘legalise’ foods that have long been forbidden is often vital for the person with bulimia. Increasing food intake bit by bit, day by day, may be a better approach for an adult with anorexia nervosa than a narrow focus on restoring weight at a speed that only results in feeling completely out of control. Clear and regular communication between clinicians, patient, carers and family is also a vital and often overlooked part of treatment.
For most people, recovery goes through several stages and is characterised by steps forward and steps back. There may be longer-term complications and a sense of an ongoing vulnerability around food long after the main issues have been resolved. During the treatment process, there is often an exchange of views on what constitutes recovery - the patient’s understanding may differ significantly from that of the professionals. There can be other issues - views on managing treatment may be at odds, understanding and dealing with the expectations of others can be very hard at times, particularly as those with eating disorders tend to want to please and have a deep internal need for external approval and affirmation. An additional dilemma may occur if the individual senses that the treatment approach is not working, whilst at the same time knowing it might be all that is on offer.
Staying positive and maintaining commitment, learning how to deal with lapses and managing the ambivalence that can flare up with a vengeance throughout recovery will be all too familiar for those who have been through this experience. The path is not always easy. But throughout this time, it is important to stay focused and to know and remember that recovery is possible and there are people who want to help overcome the disorder.
And what about those who want to help the person with an eating disorder – partners, spouses, friends, relatives, colleagues, siblings, parents? It seems that their experience is also one of complexity, ambivalence and contradiction. How can they help you face issues and accept that there is a problem? What role should they play? Where can they get information about treatment approaches and how do they communicate this? How do they deal with the medical professionals? And what about issues of patient confidentiality? Many who have been involved with somebody struggling with eating distress, have reported how desperate and isolated they have felt at times, how they have been ‘at their wits end, not knowing where to turn for help’. They will also often speak of their guilt and remorse and feeling of
45 CONFIDENTIAL Second draft for consultation September 2003
responsibility for the illness, have they caused or contributed to the situation? What could they have done to make things different?
For carers as for sufferers, the expectations of treatment; its duration and impact on those involved, may differ greatly from the actual reality. From the outset, it is important to understand and expect an arduous, challenging and unpredictable journey, with emotions previously controlled and contained, spilling out into the open.
Regardless of their nature, relationships are likely to be permanently changed by the experience of coping with an eating disorder. In the early days when the ‘sufferer’ is either stuck in denial or does not know how to even admit there is an issue, the impact on relationships can be very negative. Once the issue is out in the open things may change as the ‘sufferer’ becomes more accepting of care and support, although the road is still pretty rocky at times. Support groups of various kinds can be an important component of the healing process.
A readiness to change and to accept self responsibility and a real desire for a life where eating does not dominate every moment, are perhaps the true keys to full recovery. The experience of an eating disorder may leave a lifelong mark, however letting go, developing self-acceptance, forgiving events in the past, and dropping feelings of shame can lead to growth and to an end which in so many ways is, of course, only the beginning.
2.10.1 Clinical practice recommendations
2.10.1.1 Healthcare professionals should acknowledge the ambivalence of many people with eating disorders about treatment and the consequent demands and challenges this presents for helping them. (C)
2.10.1.2 People with eating disorders seeking help should be assessed and receive treatment at the earliest opportunity. (C)
2.10.1.3 Patients and carers should be provided with appropriate education and information on the nature, course and treatment of eating disorders. (C)
2.11 The involvement of family members / carers
2.11.1 Involving carers Eating disorders have a social as well as an individual context. Once ‘out in the open’, family members and carers are inevitably drawn into the illness and many will want to offer whatever help they can. The UK government has recognised the important role and contribution that carers provide for people
46 CONFIDENTIAL Second draft for consultation September 2003
with a range of mental health problems and information and guidance is available through a website. (www.carers.gov.uk) (Department of Health 2000, 2002b).
2.11.2 Confidentiality Some patients will be concerned that the involvement of their relatives may breach their rights of confidentiality. Consultations between those with eating disorders and healthcare professionals are bound by generic rules regarding confidentiality i.e., that this should only be breached if the patient or others are at significant risk and that a breach of confidentiality is likely to reduce that risk (Department of Health, 2002b, GMC, 2003). Furthermore the patient should be informed of any necessary breach of confidentiality. However issues of good practice with regard to confidentiality should not be accepted as an excuse for not listening to or communicating effectively with carers. Carers should be given sufficient information by medical and mental health services in a way that they can readily understand, to help them provide care effectively (Department of Health, 2003). Information from carers is also subject to the same rules of confidentiality as those applied to the individual with the illness.
2.11.3 Information for carers Carers should be given the opportunity to ask healthcare professionals about the illness in general and specific risks. The following information might reasonably be expected to be available to them:
Information about Eating Disorders • The general causes of an eating disorder
• The general maintaining factors of an eating disorder
• The best strategies to help someone with an eating disorder
• The evidence base for the treatment and management of people with anorexia nervosa
• The prognosis and expected course and outcome of people with eating disorders.
Information about physical risk • That a physical and psychological risk assessment of the patient is being regularly undertaken
• That they will be informed if the threat to the health and safety of the patient is severe
47 CONFIDENTIAL Second draft for consultation September 2003
• The danger signs that should to alert them to physical risk of the patient
• What they should do in the event that they are concerned about the health and safety of the patient and how they can recruit help when necessary.
2.11.4 Carer involvement in treatment The extent to which the family and/or next of kin need to be involved in treatment relates to age and developmental issues, the severity of the illness and the risk of harm. In general parents and other family members will be included in the treatment of children and adolescents and in the treatment of adults depending on their wishes and the assessment of risk or severity. In the rare cases where compulsory treatment is necessary, there is a statutory role for the next of kin enshrined in the 1983 Mental Health Act legislation. Therefore although services should aim to keep treatment confidential they also have to ensure that they practice safely and consider the needs of carers. They have a statutory obligation to let other people know if there are health and safety issues that need to be considered.
2.11.5 Carers experiences and wishes A survey of carers (Haigh & Treasure, 2003) and a focus group convened to inform the GDG, revealed a number of common themes: • Carers are concerned about a perceived variable range of experience and expertise in the management of eating disorders, amongst health service professionals
• They express a need for quality information and effective communication
• They often express a wish to be included in their relative’s ‘treatment team’
• They are concerned about a perceived lack (in many areas) of adequate service provision and many worry about delays in accessing effective services and smooth transition between tiers of care.
The issues raised can be classified as engaging with primary care, experiences of treatment, and issues in aftercare. The carer’s focus group also made the following observations.
2.11.5.1 Engaging with primary care • GPs may think issues and symptoms are trivial, partly because they often have little experience of eating disorder’s
48 CONFIDENTIAL Second draft for consultation September 2003
• A GP’s inexperience and lack of training in eating disorder can lead to a delay in diagnosis
• Pathways between primary and secondary care are slow and fraught with obstacles, including lack of choice
• There are often difficulties in receiving information and effective communication particularly in relation to the treatment of older adolescents. These relate both to issues of confidentiality, and to problems in communication in the transition between child and adolescent and adult services.
2.11.5.2 Experiences of treatment • Concern was expressed about the poor response of adolescents to inpatient treatment for a range of reasons
• Poor communication is common, partly related to shift working in inpatient units. The treatment plan was sometimes poorly explained.
• Carers need information about the disorders, their own rights, and compulsory treatment.
• Siblings need support as well as parents / adult carers.
• Support, for example from relatives support groups was welcomed
• Community based services are preferred to inpatient treatment where this can be managed.
2.11.5.3 Aftercare • Aftercare provision is often limited and can undermine the benefit of inpatient treatment
• Responsibility for aftercare is often unclear
• Co-existing or residual difficulties are often not addressed
• Little choice is offered.
2.11.5.4 Carer advice The focus group suggested that improvements could be made in the following areas:
49 CONFIDENTIAL Second draft for consultation September 2003
• More training of health service staff is required, particularly for those in general practice including GPs. This would include training in the detection and the features of eating disorders and an understanding of how motivation and ambivalence can affect engagement in treatment
• Carers should be actively involved in treatment programmes
• Treatment plans and other information should be provided to carers, including written reference resources
• Carers should be offered the opportunity to learn from and benefit from the experience and support of others in the same situation
• Recovered or recovering patients could be used to provide support to patients at an earlier stage of treatment
• A greater focus on outreach and community services rather than inpatient services is desirable, in order to avoid dislocation from the patient’s family and social support
• A greater uniformity of treatment protocols and approaches, based on the available evidence rather than service constraints is desirable, in order to reduce idiosyncratic practice.
2.11.5.5 In addition to the provision of high quality information, family and carers may be informed of self-help groups and support groups and offered the opportunity to participate in such programmes where they exist. (C)
2.12 Stigma The above review of the experience of eating disorders from the patients' and carers’ perspective reveals the disabling nature of these disorders. Those who suffer from eating disorders often carry the added burden of stigmatising attitudes from the lay public and health services staff. These attitudes not only restrict the opportunities for effective treatment, but also confer additional handicaps. The Royal College of Psychiatrists stigma campaign (Cowan & Hart, 1998), followed a survey of the general public which revealed that stigmatising attitudes were based on four beliefs about those with mental health problems, namely:
• Sufferers are thought to be dangerous to others
• The disorders are thought to be self-inflicted
• The outcome is thought to be poor
50 CONFIDENTIAL Second draft for consultation September 2003
• It is difficult to communicate with sufferers.
Crisp et al. (2000) have indeed shown that those with eating disorders are believed by the general public to be unpredictable, hard to talk to, are different from them, have only themselves to blame and should ‘pull themselves together’. Gowers and Shore (1999) have reviewed the part played by these factors in stigmatising those with eating disorders and have argued for improved training of health service staff and greater public education. Crucially, stigma can be reduced by placing the person with an eating disorder at the centre of their treatment and by engaging them in a positive therapeutic relationship (see section 2.9). This is a key aim of this guideline.
51 CONFIDENTIAL Second draft for consultation September 2003
3 Methods used to develop this guideline
3.1 Overview The development of this guideline drew upon methods outlined by NICE (NICE, 2002; Eccles & Mason, 2001). A team of experts, professionals, and a patient representative, known as the Guideline Development Group (GDG), undertook the development of a patient centred, evidence-based guideline with support from the NCCMH staff. There are five basic steps in the process of developing a guideline:
• Define clinical questions considered important for practitioners and patients
• Develop criteria for evidence searching and search for evidence
• Design validated protocols for systematic review and apply to evidence recovered by search
• Synthesise and (meta-) analyse data retrieved, guided by the clinical questions
• Answer clinical questions with evidence-based recommendations for clinical practice.
The clinical practice recommendations made by the GDG are therefore derived from the most up-to-date and robust evidence base for the clinical and cost effectiveness of the treatments and services used in the management of eating disorders. In addition, to ensure a patient and carer focus, their concerns regarding clinical practice have been highlighted and addressed by recommendations agreed by the whole GDG. The evidence-based recommendations are the core of this guideline.
3.2 The Guideline Development Group The Eating Disorders GDG consisted of: professionals in psychiatry, clinical psychology, nursing, social work, and general practice; academic experts in psychiatry and psychology; a patient, and a representative from a patient organisation. The carer perspective was provided through focus group discussion with carers; the group was run by the patient on the GDG. The guideline development process was supported by staff from the NCCMH
52 CONFIDENTIAL Second draft for consultation September 2003 review team, who undertook the clinical and health economics literature searches, reviewed and presented the evidence to the GDG, managed the process, and contributed to the drafting of the guideline.
3.2.1 Guideline Development Group meetings Twenty three eating disorders GDG meetings were held between March 2002 and October 2003. During the series of day-long meetings, clinical questions were written, clinical evidence was reviewed and assessed, statements developed and recommendations formulated. At each meeting, all GDG members declared any potential conflict of interests, and patient and carer concerns were routinely discussed as part of a standing agenda.
3.2.2 Topic groups The GDG divided its workload along clinically relevant lines to simplify the guideline development process by forming smaller topic groups to undertake guideline work in specified areas of clinical practice. Topic groups covered physical management, service-level interventions, and psychological interventions. These groups were designed to efficiently manage the large volume of evidence appraisal prior to presenting it to the GDG as a whole. Each topic group was chaired by a GDG member with expert knowledge of the topic area (one of the healthcare professionals). Topic groups refined the clinical definitions of treatment interventions, identified relevant clinical questions, reviewed and prepared the evidence with the systematic reviewer before presenting it to the GDG as a whole, and helped the GDG to identify further expertise in the topic. Topic group leaders reported the status of the group’s work as part of the standing agenda. They also introduced and led the GDG discussion of the evidence review for that topic and assisted the GDG Chair in drafting that section of the guideline relevant to the work of each topic group.
3.2.3 Patients and carers The GDG included a patient and representatives of a national patient group. Given their direct experience of services, they gave an integral patient focus to the GDG and the guideline. They contributed as full GDG members to writing the clinical questions, helping to ensure that the evidence addressed their views and preferences, highlighting sensitive issues and terminology associated with eating disorders, and bringing service-user research to the attention of the GDG. They wrote the section on the patient perspective, organised, conducted and reported on the carer focus group contribution, and identified recommendations from the patient and carer perspective.
The carer perspective was provided through a focus group held with carers. The main objective of the group discussion was to discuss carers views about communication with clinicians and other professionals, including guidance
53 CONFIDENTIAL Second draft for consultation September 2003
on responsibility and expectations of clinicians, carers, and people with eating disorders; the role of carers in identifying eating disorders and subsequent treatment, and accessing appropriate services. Twelve people, currently caring for someone with anorexia nervosa as a parent or spouse were recruited by the Eating Disorders Association and the Carers volunteers Database at the Eating Disorders Research Unit of the Institute of Psychiatry. Geographic distribution of members included Cornwall, the Midlands, East Anglia, southeast England and London. Participants were told the purpose of the session, and signed a participant consent form. Discussions were tape- recorded with the full knowledge of the participants, transcribed, and key themes identified.
3.2.4 Special advisors Special advisors who had specific expertise in one or more aspects of the treatment and management of eating disorders were invited to assist the GDG, to comment on specific aspects of the developing guideline and to present evidence and observations on areas of their expertise to the GDG. Appendix 2 lists those who agreed to act as special advisors.
3.2.5 National and international experts National and international experts in eating disorders research were identified through the literature search and through the experience of the GDG members. These experts were invited to recommend unpublished or soon-to-be published studies in order to ensure that the most recent evidence on the management of eating disorders was included in the development of the guideline. They informed the group about completed trials at the pre- publication stage, systematic reviews in the process of being published, studies relating to the cost-effectiveness of treatment, and trial data if the GDG could be provided with full access to the complete trial report. Appendix 5 lists researchers who were contacted.
3.3 Clinical questions Clinical questions were used to guide the identification and interrogation of the evidence base relating to the use of psychological interventions, service- level interventions, and physical interventions in the treatment and management of eating disorders. The GDG necessarily had to limit the number of questions to those they regarded as important. Questions were developed for each clinical topic area. Appendix 6 lists the clinical questions.
3.4 Systematic clinical literature review strategies The aim of the clinical literature review was to systematically identify and synthesise relevant evidence from the literature in order to answer specific clinical questions developed by the GDG. Thus, clinical practice
54 CONFIDENTIAL Second draft for consultation September 2003
recommendations are evidence-based, where possible, and if evidence is not available, gaps are identified where future research is needed.
3.4.1 Methodology A stepwise, hierarchical approach was taken to locating and presenting evidence to the GDG. The NCCMH review team had developed this process based on advice from the National Guidelines Support and Research Unit (NICE) and after considering recommendations from a range of other sources. These included: • Centre for Clinical Policy and Practice of the New South Wales Health Department (Australia)
• Clinical Evidence Online
• Cochrane Collaboration
• New Zealand Guideline Group
• NHS Centre for Reviews and Dissemination
• Oxford Centre for Evidence-Based Medicine
• Scottish Intercollegiate Guidelines Network (SIGN)
• United States Agency for Health Research and Quality
• Oxford Systematic Review Development Programme.
3.4.2 The review process
3.4.2.1 Questions of treatment efficacy For questions related to the efficacy of treatment, the initial evidence base was formed from high-quality, recently published or updated randomised controlled trials (RCTs) that addressed at least one of the clinical questions developed by the GDG. Systematic reviews were selected on predetermined quality criteria. Further searches for new RCTs were undertaken. New RCTs meeting inclusion criteria set by the GDG were incorporated into existing systematic reviews and fresh analyses performed. If no systematic reviews were available, the review team located all relevant high quality RCTs for review and, where appropriate, meta-analysis. The review process is illustrated in Flowchart 1.
55 CONFIDENTIAL Second draft for consultation September 2003
Flowchart 1: Eating disorders Guideline Review Process
Key Identify clinical questions
Conduct systematic search for relevant Start Action Dec is ion systematic review s (initial 5 year limit) required to be taken No >1 located?
Yes Repeat process searching Scan titles and abstracts & apply eligibility for next level of evidence criteria liberally (e.g., RCTs )
Cross-check excluded citations; if any doubt, retrieve full paper Number of citations excluded Retrieve all eligible papers for more Check for detailed evaluation duplicates Number of citations that could not be located
Re-evaluate eligibility & assess quality using tw o independent coders, cross-check results for reliability List papers excluded w ith reasons
Conduct systematic search for No >1 meets both eligibility & appropriate level of evidence quality criteria?
Retrieve evidence & assess Yes eligibility/quality (use 2 independent review ers)
Cross-check results for >1 meets both No reliability criteria? Eligibility/quality Yes assessment process checked by appropriate Topic Group Extract data; cross-check results for reliability No New RCT data Yes searched for? Enter data into RevMan (contact author if results not reported in sufficient detail)
Synthesise results (using meta-analytical techniques w here appropriate & perform List papers excluded from sensitivity analyses) meta-analysis (but still included in narrative analysis) w ith reasons Construct evidence tables & present to Guideline Development Group
GDG generates draft statements & recommendations
Statements & recommendations review ed in light of economic evidence
56 CONFIDENTIAL Second draft for consultation September 2003
Although there are a number of difficulties with the use of RCTs in the evaluation of interventions in mental health, some of which also apply to the use of RCTs in any health research, the RCT remains the most important method for establishing efficacy. However, in some cases it was not possible to identify high-quality systematic reviews or a substantial body of RCTs that directly addressed a clinical question. In this situation, an informal consensus process was adopted (see Section 3.4.7). Future guidelines on the treatment and management of eating disorders will be able to update and extend the usable evidence base starting from the evidence collected, synthesised and analysed for this guideline.
3.4.2.2 Questions of diagnosis and prognosis For questions related to diagnosis and prognosis, the initial evidence base was formed from studies with the most appropriate and reliable design to answer the particular question (i.e., diagnosis – cross sectional studies, prognosis – cohort studies of representative patients). The review process was similar to that described for questions of efficacy. The main difference concerned the study search filter and the criteria used to assess methodological quality. In situations where it was not possible to identify high-quality systematic reviews or a substantial body of appropriately designed studies that directly addressed a clinical question, an informal consensus process was adopted (see Section 3.4.7).
3.4.3 Search strategies In conducting the review, the team systematically searched the literature for all English-language systematic reviews relevant to the eating disorders scope that were published or updated after 1995.
Search filters developed by the review team consisted of a combination of subject heading and free-text phrases. A general filter was developed for eating disorders along with more specific filters for each clinical question. In addition, filters were developed for RCTs and for other appropriate research designs. (The search filters can be found in Appendix 8.)
Electronic searches were made of the major bibliographic databases (MEDLINE, EMBASE, PsycINFO, CINAHL), in addition to the Cochrane Database of Systematic Reviews, the NHS R&D Health Technology Assessment database, Evidence-Based Mental Health, and Clinical Evidence (Issue 5).
Ineligible articles were excluded, and a second independent reviewer crosschecked these for relevance. The remaining references were acquired in full and re-evaluated for eligibility. The most recently published reviews that appropriately addressed a clinical question were selected. For each systematic
57 CONFIDENTIAL Second draft for consultation September 2003
review used, a search was made for new studies, and the papers for these and for existing studies, were retrieved.
The search for further evidence included research published after each review’s search date, in-press papers identified by experts, and reviewing reference lists and recent contents of selected journals. All reports that were retrieved but later excluded are listed with reasons for exclusion in the appropriate evidence table. Where no relevant systematic reviews were located, the review team asked the GDG to decide whether a fresh systematic review should be undertaken. Eligible reviews were critically appraised for methodological quality and the reliability of this procedure was confirmed by parallel independent assessment. The eligibility/quality assessment was tested on a representative sample of papers. (Appendix 10 provides the quality checklist.)
3.4.4 Synthesising the evidence Where possible, outcome data were extracted directly from all eligible studies that met the quality criteria into Review Manager 4.2 (Cochrane Collaboration, 2003). Meta-analysis was then used to synthesise the evidence where appropriate using Review Manager. If necessary, reanalyses of the data or sensitivity analyses were used to answer clinical questions not addressed in the original studies or reviews. Where meta-analysis was not appropriate and/or possible, the reported results from each primary-level study were entered into the Access database. Evidence tables were used to summarise general information about each study. Consultation was used to overcome difficulties with coding. Data from studies included in existing systematic reviews were extracted independently by one reviewer directly into Review Manager and crosschecked with the existing data set. Two independent reviewers extracted data from new studies, and disagreements were resolved by discussion. Where consensus could not be reached, a third reviewer resolved the disagreement. Masked assessment (i.e., blind to the journal from which the article comes, the authors, the institution, and the magnitude of the effect) was not used since it is unclear that doing so reduces bias (Jadad et al., 1996; Berlin, 2001).
3.4.5 Presenting the data to the GDG Where possible, the GDG was given a graphical presentation of the results using forest plots generated with the Review Manager software. Each forest plot displayed the effect size and confidence interval (CI) for each study as well as the overall summary statistic. The graphs were organised so that the display of data in the area to the left of the ‛line of no effect’ indicated a ‛favourable’ outcome for the treatment in question. Dichotomous outcomes were presented as relative risks (RR) and the associated 95% CI (see Figure 1). A relative risk (or risk ratio) is the ratio of the treatment event rate to the control event rate. A RR of 1 indicates no difference between treatment and
58 CONFIDENTIAL Second draft for consultation September 2003
control. In Figure 1, the overall RR of 0.73 indicates that the event rate (i.e., non-remission rate) associated with intervention A is about ¾ of that with the control intervention, or in other words, intervention A reduces non-remission rates by 27%. In addition, the CI around the RR does not cross the ‘line of no effect’ indicating that this is a statistically significant effect. The CI shows with 95% certainty the range within which the true treatment effect should lie.
All dichotomous outcomes were calculated on an intention-to-treat basis (i.e., a ‛once-randomised-always-analyse’ basis). This assumes that those participants who ceased to engage in the study – from whatever group – had an unfavourable outcome (with the exception of the outcome of ‛death’). The Number Needed to Treat (NNT) or the Number Needed to Harm (NNH) was reported for each statistically significant outcome where the baseline risk (i.e., control group event rate) was similar across studies. In addition, NNTs calculated at follow-up were only reported where the length of follow-up was similar across studies. When length of follow-up or baseline risk varies (especially with low risk), the NNT is a poor summary of the treatment effect (Deeks, 2002).
Continuous outcomes were analysed as weighted mean differences (WMD), or as standardised mean differences (SMD) when different measures were used in different studies to estimate the same underlying effect (see Figure 2).
Review: NCCMH clinical guideline review (Example) Comparison: 01 Intervention A compared to a control group Outcome: 01 Number of people who did not show remission
Study Intervention A Control RR (fixed) Weight RR (fixed) or sub-category n/N n/N 95% CI % 95% CI
01 Intervention A vs. control Griffiths1994 13/23 27/28 38.79 0.59 [0.41, 0.84] Lee1986 11/15 14/15 22.30 0.79 [0.56, 1.10] Treasure1994 21/28 24/27 38.92 0.84 [0.66, 1.09] Subtotal (95% CI) 45/66 65/70 100.00 0.73 [0.61, 0.88] Test for heterogeneity: Chi² = 2.83, df = 2 (P = 0.24), I² = 29.3% Test for overall effect: Z = 3.37 (P = 0.0007)
0.2 0.5 1 2 5 Favours intervention Favours control Figure 1. Example of a forest plot displaying dichotomous data
Review: NCCMH clinical guideline review (Example) Comparison: 01 Intervention A compared to a control group Outcome: 03 Mean frequency (endpoint)
Study Intervention A Control SMD (fixed) Weight SMD (fixed) or sub-category N Mean (SD) N Mean (SD) 95% CI % 95% CI
01 Intervention A vs. control Freeman1988 32 1.30(3.40) 20 3.70(3.60) 25.91 -0.68 [-1.25, -0.10] Griffiths1994 20 1.25(1.45) 22 4.14(2.21) 17.83 -1.50 [-2.20, -0.81] Lee1986 14 3.70(4.00) 14 10.10(17.50) 15.08 -0.49 [-1.24, 0.26] Treasure1994 28 44.23(27.04) 24 61.40(24.97) 27.28 -0.65 [-1.21, -0.09] Wolf1992 15 5.30(5.10) 11 7.10(4.60) 13.90 -0.36 [-1.14, 0.43] Subtotal (95% CI) 109 91 100.00 -0.74 [-1.04, -0.45] Test for heterogeneity: Chi² = 6.13, df = 4 (P = 0.19), I² = 34.8% Test for overall effect: Z = 4.98 (P < 0.00001)
-4 -2 0 2 4 Favours intervention Favours control Figure 2. Example of a forest plot displaying continuous data
Both the I2 test of heterogeneity and the chi-squared test of heterogeneity (p < .10) were used, as well as visual inspection of the forest plots, to look for the
59 CONFIDENTIAL Second draft for consultation September 2003
possibility of heterogeneity. I2 describes the proportion of total variation in study estimates that is due to heterogeneity (Higgins & Thompson, 2002). An I2 of less than 30% was taken to indicate mild heterogeneity and a fixed effects model was used to synthesise the results. An I2 of more than 50% was taken as notable heterogeneity. In this case an attempt was made to explain the variation. If studies with heterogeneous results were found to be comparable, a random effects model was used to summarise the results (DerSimonian & Laird, 1986). In the random effects analysis, heterogeneity is accounted for both in the width of CIs and in the estimate of the treatment effect. With decreasing heterogeneity the random effects approach moves asymptotically towards a fixed effects model. An I2 of 30 to 50% was taken to indicate moderate heterogeneity. In this case, both the chi-squared test of heterogeneity and a visual inspection of the forest plot was used to decide between a fixed and random effects model.
To explore the possibility that the results entered into each meta-analysis suffered from publication bias, data from included studies were entered, where there was sufficient data, into a funnel plot. Asymmetry of the plot was taken to indicate possible publication bias and investigated further.
3.4.6 Forming and grading the statements and recommendations The evidence tables and forest plots formed the basis for developing clinical statements and recommendations. For intervention studies, the statements were classified according to an accepted hierarchy of evidence. Recommendations were then graded A to C based on the level of associated evidence (see text box 1). Key recommendations are denoted by an asterisk following the grade.
In order to facilitate consistency in generating and drafting the clinical statements the GDG utilised a statement decision tree (see Flowchart 2). The flowchart was designed to assist with, but not replace clinical judgement.
Where a statistically significant summary statistic (effect size; ES) was obtained (after controlling for heterogeneity), the GDG considered whether this finding was of clinical significance (i.e., likely to be of benefit to patients) taking into account the trial population, nature of the outcome and size of the effect. On the basis of this consideration the ES was characterised as ‛clinically significant’ or not. A further consideration was made about the strength of the evidence by examining the CI surrounding the ES. For level I evidence, where the ES was judged to be clinically significant and had a CI entirely within a clinical relevant range, the result was characterised as ‛strong evidence’ (S1, Flowchart 2). For non-level I evidence or in situations where the upper/lower bound of the CI was not clinically significant, the result was characterised as ‛limited evidence’ (S2). Where an ES was statistically significant, but not clinically significant and the CI excluded values judged to be clinically important, the result was characterised as ‘unlikely to be clinically significant’
60 CONFIDENTIAL Second draft for consultation September 2003
(S3). Alternatively, if the CI included clinically important values, the result was characterised as ‘insufficient to determine clinical significance’ (S6).
Text Box 1: Hierarchy of evidence and recommendations grading scheme
Level Type of evidence Grade Evidence
I Evidence obtained from a single A At least one randomised controlled randomised controlled trial or a meta- trial as part of a body of literature analysis of randomised controlled of overall good quality and trials consistency addressing the specific recommendation (evidence level I) without extrapolation
IIa Evidence obtained from at least one B Well-conducted clinical studies but well-designed controlled study no randomised clinical trials on the without randomisation topic of recommendation (evidence levels II or III); or extrapolated from level I evidence
IIb Evidence obtained from at least one other well-designed quasi- experimental study
III Evidence obtained from well- designed non-experimental descriptive studies, such as comparative studies, correlation studies and case studies
IV Evidence obtained from expert C Expert committee reports or committee reports or opinions and/or opinions and/or clinical clinical experiences of respected experiences of respected authorities authorities (evidence level IV). This grading indicates that directly applicable clinical studies of good quality are absent or not readily available
NICE Evidence from NICE guideline or NICE Evidence from NICE guideline or health technology appraisal health technology appraisal
Adapted from Eccles, M. & Mason, J. (2001). How to develop cost-conscious guidelines. Health Technology Assessment 5: 8; NHS Executive. Clinical Guidelines: Using Clinical Guidelines to Improve Patient Care Within the NHS. London: 1996.
61 CONFIDENTIAL Second draft for consultation September 2003
Flowchart 2: Guideline Statement Decision Tree
Start Key
Start Decision Action to be taken required
Statistical CI = Confidence Interval difference ES = Ef f e ct Si z e between x and y NO (after controlling for heterogeneity)?
CI completely excludes NO YES clinically important ESs?
YES Clinical NO significance? S4
"There is evidence YES suggesting that there is no clinically significant difference between x and y on… " CI NO only contains clinically important ESs? S5
"There is insufficient evidence to YES determine if there is a clinically significant difference between x and y on…" NO Level I CI completely evidence? excludes NO clinically important ESs?
YES S2 YES S6 "There is limited evidence suggesting "There is evidence that there is a suggesting that there clinically significant is a statistically difference between x significant difference between x and y but S1 and y with x being superior on…" there is insufficient evidence to "There is strong determine its clinical evidence suggesting significance." that there is a S3 clinically significant difference between x "There is evidence and y with x being suggesting that there superior on…" is a statistically significant difference between x and y but the size of this difference is unlikely to be of clinical significance."
62 CONFIDENTIAL Second draft for consultation September 2003
Where a non-statistically significant ES was obtained, the GDG reviewed the trial population, nature of the outcome, size of the effect and, in particular, the CI surrounding the result. If the CI was narrow and excluded a clinically significant ES, this was seen as indicating evidence of ‘no clinically significant difference’ (S4), but where the CI was wide this was seen as indicating ‘insufficient evidence’ to determine if there was a clinically significant difference or not (S5).
Once all evidence statements relating to a particular clinical question were finalised and agreed by the GDG, the associated recommendations were produced and graded. Grading the recommendations allowed the GDG to distinguish between the level of evidence and the strength of the associated recommendation. It is possible that a statement of evidence would cover only one part of an area in which a recommendation was to be made or would cover it in a way that would conflict with other evidence. In order to produce more comprehensive recommendations suitable for people in England and Wales, the GDG had to extrapolate from the available evidence. This led to a weaker level of recommendation (i.e., B, as data were based upon level I evidence). In addition, it is possible to have methodologically sound (level I) evidence about an area of practice that is of little direct clinical relevance or has such a small effect that it is of little practical importance. In this case, the evidence would attract a lower strength of recommendation (i.e., there would be necessity for extrapolation). It is important to note that the grading of the recommendation is not a reflection of its clinical significance or relevance.
A number of issues relating to the study of eating disorders (see section 3.4.8) meant that the outcomes available for analysis were classified as primary or secondary. When making recommendations, the primary outcomes were given more weight during the decision process.
The process also allowed the GDG to moderate recommendations based on factors other than the strength of evidence. Such considerations include the applicability of the evidence to people with eating disorders, economic considerations, values of the development group and society, or the group’s awareness of practical issues (Eccles et al., 1998).
3.4.7 Method used to answer a clinical question in the absence of appropriately designed, high-quality research Where it was not possible to identify at least one appropriately designed study or high-quality systematic review, or where the GDG were of the opinion (on the basis of previous searches or their knowledge of the literature) that there were unlikely to be appropriately designed primary-level research that directly addressed the clinical question, an informal consensus process was adopted. This process focused on those questions that the GDG considered a priority.
63 CONFIDENTIAL Second draft for consultation September 2003
The starting point for this process of informal consensus was that a member of the topic group identified, with help from the systematic reviewer, a narrative review that most directly addressed the clinical question. Where this was not possible a new review of the recent literature was initiated.
This existing narrative review or new review was used as a basis for identifying lower levels of evidence relevant to the clinical question. This was then presented for discussion to the GDG. On the basis of this, additional information was sought and added to the information collected. This may include studies that did not directly address the clinical question but were thought to contain relevant data. This led to the development of an initial draft report that addressed the following issues:
• A description of what is known about the issues concerning the clinical question
• A brief review of the existing evidence, including RCTs, non- randomised controlled studies, cohort studies and other studies that help answer the question
• The summary of the evidence so far obtained. This was then presented in narrative form to the GDG and further comments were sought about the evidence and its perceived relevance to the clinical question
• If, during the course of preparing the report, a significant body of primary-level studies (of appropriate design to answer the question) were identified, a full systematic review was done
• At this time, subject possibly to further reviews of the evidence, a series of statements that directly addressed the clinical question were developed
• Following this, on occasions and as deemed appropriate by the development group, the report was then sent to appointed experts outside of the GDG for peer review and comment. The information from this process was then fed back to the GDG for further discussion of the statements
• Recommendations were then developed and could also be sent for further external peer review
• After this final stage of comment, the statements and recommendations were again reviewed and agreed upon by the GDG.
64 CONFIDENTIAL Second draft for consultation September 2003
3.4.8 Issues concerning research on eating disorders Studying the treatment of eating disorders presents certain particular challenges. First, since psychological treatments have a major role in their management, the methods needed to evaluate psychotherapeutic treatments have to be adopted. Some of these are mentioned in the national clinical practice guideline on schizophrenia (Kendall et. al., 2003) including the need for some degree of standardisation of the psychological treatment provided (usually in the form of a treatment manual) and the importance of quality assessment and control. Additional refinements have been included in some eating disorder studies including the use of “dismantling” research designs to identify the active components of psychological treatments (e.g., Fairburn et al., 1991, 1993); studies of the mediators of psychological treatments, a stage towards the identification of their mechanism of action (e.g., Wilson et al., 2002); fine-grain comparisons of different ways of implementing specific psychotherapeutic procedures (e.g., Bulik et al., 1998); and “effectiveness” studies of simplified psychological treatments designed for general dissemination (e.g., Carter & Fairburn, 1998).
The second challenge is the measurement of outcome. Eating disorders affect many aspects of functioning including behaviour (e.g., eating habits), attitudes and values (e.g., the evaluation of shape and weight), psychosocial functioning (e.g., levels of depression and anxiety; interpersonal functioning) and physical health (e.g., body weight, serum electrolytes) (Fairburn & Harrison, 2003). It is also usual for there to be comorbidity with Axis I and II disorders co-existing with the eating disorder (Bulik, 2002). Therefore, most treatment studies employ a range of different outcome measures to determine whether the treatments studied affect all or part of the psychopathology present. Having said this, there are accepted primary outcome variables that treatments must influence if they can be said to have an impact on the core eating disorder. For anorexia nervosa, the primary outcome variable is body weight adjusted for height, usually represented as the BMI or the percentage of expected weight for the person’s age, height and sex. For bulimia nervosa the comparable variable is the frequency of binge eating and ‘purging’ (self- induced vomiting or the use of laxatives to influence body shape or weight); that is, the frequency of these forms of behaviour over a set period of time (usually one or four weeks). In addition, it is now established practice to report the proportion of participants who no longer practise the behaviour (sometime referred to as the “abstinence” rates). For binge eating disorder the primary outcome variable is the frequency of binge eating, represented as for bulimia nervosa. An important point to note is that the frequency of binge eating and purging is skewed rather than being normally distributed so these data have to be transformed prior to analysis. However, because most reports give uncorrected means and standard deviations, the GDG gave less weight to frequency data than abstinence rates when making recommendations.
65 CONFIDENTIAL Second draft for consultation September 2003
The third challenge is determining whether treatment effects persist over time. Anorexia nervosa and bulimia nervosa tend to run a chronic course so short-lived treatment effects are of limited clinical significance. Thus it has become established research practice to follow-up patients following the end of treatment to see whether any changes last, and with interventions that are known to be associated with a high rate of relapse (e.g., inpatient treatment of anorexia nervosa) there have been attempts to develop and evaluate subsequent “relapse prevention” strategies (e.g., Kaye et al., 2001; Pike et al., 2003).
The fourth challenge is largely peculiar to research on the treatment of anorexia nervosa. It is exceptionally difficult recruiting cases for treatment studies. This is for a number of reasons. First, anorexia nervosa is uncommon. Second, it can be life-threatening with the result that it is not be ethical to recruit certain patients. And third, patients with anorexia nervosa are notoriously reluctant to have treatment and this reluctance can be magnified if the clinician raises the possibility of taking part in a research trial. For these reasons the few studies of the treatment of anorexia nervosa have mostly involved small numbers of patients. As a result it is difficult to draw robust evidence-based conclusions about the treatment of the disorder.
3.5 Health economics review strategies The aim of the health economics review was to contribute to the guideline development process data on the economic burden of eating disorders and evidence of cost-effectiveness of the different treatment options for eating disorders were collected and assessed in order to help the decision-making process. See Chapter 9, Health Economics Evidence, for the detailed review strategies.
3.6 Stakeholder contributions Professionals, patients, and companies have contributed to and commented on the eating disorders guideline at key stages in its development. Stakeholders for this guideline include:
• Patient/carer stakeholders: the national patient and carer organisations that represent people whose care is described in this guideline
• Professional stakeholders: the national organisations that represent healthcare professionals who are providing services to patients
• Commercial stakeholders: the companies that manufacture medicines used in the treatment of eating disorders
• Primary Care Trusts
66 CONFIDENTIAL Second draft for consultation September 2003
• Department of Health and Welsh Assembly Government.
Stakeholders have been involved in the guideline’s development at the following points:
• Commenting on the initial scope of the guideline with patient organisation stakeholders attending a briefing meeting held by NICE
• Contributing lists of evidence to the GDG
• Commenting on the first and second drafts of the guideline.
3.7 Validation of this guideline This guideline has been validated through two consultation exercises. The first consultation draft was submitted to the NICE Guidelines Review Panel, and circulated to stakeholders, special advisors, and other reviewers nominated by GDG members.
After taking into account comments from stakeholders, the NICE Guidelines Review Panel, a number of health authority and trust representatives and a wide range of national and international experts from this round of consultation, the GDG responded to all comments and prepared a final consultation draft which was submitted to NICE, circulated to all stakeholders for final comments and posted on the website for public consultation. The final draft was then submitted to the NICE Guidelines Review Panel for review prior to publication.
67 CONFIDENTIAL Second draft for consultation September 2003
4 Summary of recommendations
Key recommendations
Anorexia nervosa • Most adults with anorexia nervosa should be managed on an outpatient basis with psychological treatment provided by a healthcare professional competent in the psychological treatment and assessment of physical risk of people with eating disorders
• Patients with anorexia nervosa requiring inpatient treatment should normally be admitted to a setting that can provide the skilled implementation of refeeding with careful physical monitoring and in combination with psychosocial interventions
• Family interventions focused explicitly on eating disorders should be offered to children and adolescents with anorexia nervosa.
Bulimia nervosa • Cognitive behaviour therapy for bulimia nervosa (CBT-BN), a specifically adapted form of CBT, should be offered to most adults with bulimia nervosa
• As an alternative or additional first step to using an evidence-based self-help programme, adults with bulimia nervosa may be offered a trial of an antidepressant drug
• Older adolescents with bulimia nervosa may be treated with CBT-BN adapted as needed to suit their age, circumstances and level of development.
Atypical eating disorders • In the absence of evidence to guide the management of atypical eating disorders (eating disorders not otherwise specified) other than binge eating disorder, it is recommended that the clinician considers following the guidance on the treatment of the eating problem that most closely resembles the individual patient’s eating disorder
• Cognitive behaviour therapy for binge eating disorder (CBT-BED), a specifically adapted form of CBT, should be offered to adults with binge eating disorder
68 CONFIDENTIAL Second draft for consultation September 2003
The following guidance is evidence based. The grading scheme used for the recommendations (A, B, C) is described in section 3.4.6; a summary of the evidence on which the guidance is based is provided in chapters 5 through 8. 4.1 Introduction This guideline makes recommendations for the identification, treatment and management of anorexia nervosa, bulimia nervosa, and related eating disorders in primary, secondary and tertiary care.
The first section that follows contains the recommendations that apply across all three conditions. 4.2 Care across all conditions
Certain elements of care are common to all the eating disorders covered in this guideline. They include effective assessment, the coordination of care, the involvement of patients and carers and the provision of good and appropriate information to patients and carers.
4.2.1 Assessment and Coordination of Care
4.2.1.1 Assessment of people with eating disorders should be comprehensive and include medical, psychological and social needs, and a comprehensive assessment of risk to self. (C)
4.2.1.2 The level of risk to the patient’s mental and physical health should be monitored as treatment progresses as it may change – for example following weight gain or at times of transition in cases of anorexia nervosa. (C)
4.2.1.3 For people with eating disorders presenting in primary care, GPs should take responsibility for the initial assessment and the initial coordination of care. This includes the determination of the need for emergency medical or psychiatric assessment. (C)
4.2.1.4 Where management is shared between primary and secondary care, there should be clear agreement between individual healthcare professionals on the responsibility for monitoring of patients with eating disorders. This agreement should be in writing and should be shared with the patient and, where appropriate, families and carers. (C)
4.2.2 Providing good information and mutual support The provision of information and support is important in promoting understanding and collaboration between patients, their families and carers and healthcare professionals.
69 CONFIDENTIAL Second draft for consultation September 2003
4.2.2.1 Patients and carers should be provided with appropriate education and information on the nature, course and treatment of eating disorders. (C)
4.2.2.2 In addition to the provision of high quality information, family and carers may be informed of self-help groups and support groups and offered the opportunity to participate in such programmes where they exist. (C)
4.2.2.3 Healthcare professionals should acknowledge the ambivalence of many people with eating disorders about treatment and the consequent demands and challenges this presents for successful treatment. (C)
4.2.3 Getting help early In anorexia nervosa and bulimia nervosa, there can be serious long-term consequences to a delay in obtaining treatment.
4.2.3.1 People with eating disorders seeking help should be assessed and receive treatment at the earliest opportunity. (C)
4.2.3.2 Whenever possible patients should be engaged and treated before reaching severe emaciation. This requires both early identification and intervention. Effective monitoring and engagement of patients at severely low weight or with falling weight should be a priority. (C)
4.2.4 Special considerations for children and adolescents
4.2.4.1 Healthcare professionals assessing children and adolescents with eating disorders should be alert to indicators of abuse (emotional, physical and sexual) and should remain so throughout treatment. (C)
4.2.4.2 The right to confidentiality of children and adolescents with eating disorders should be respected. (C)
4.2.4.3 Healthcare professionals working with children and adolescents with eating disorders should familiarise themselves with national guidelines and their employers’ policies in the area of confidentiality. (C)
4.3 Identification and management of eating disorders in primary care and non-mental health settings
Primary care and non-specialist mental health settings are commonly where people with eating disorders first present. Healthcare professionals, in particular GPs, working in these areas need to be alert to the presence of
70 CONFIDENTIAL Second draft for consultation September 2003
eating disorders and take appropriate steps to screen and identify individuals with eating disorders.
4.3.1.1 Target groups for screening questions should include young women with low body mass index (BMI) compared with age norms, patients consulting with weight concerns who are not overweight, menstrual disturbances or amenorrhoea, gastrointestinal symptoms, patients with physical signs of starvation or repeated vomiting and children with poor growth. (C)
4.3.1.2 When screening for eating disorders one or two simple questions should be considered for use with specific target groups (for example, ”Do you think you have an eating problem?” and ”Do you worry excessively about your weight?”) (C)
4.3.1.3 Young women with Type 1 Diabetes and poor treatment adherence should be screened and assessed for the presence of an eating disorder. (C)
4.3.1.4 In anorexia nervosa, although weight and body mass index (BMI) are important indicators of physical risk they should not be considered the sole indicators (as on their own they are unreliable in adults and especially in children). (C)
4.3.1.5 In assessing whether a person has anorexia nervosa, attention should be paid to the overall clinical assessment (repeated over time), including rate of weight loss, growth rates in children, objective physical signs and appropriate laboratory tests. (C)
4.3.2 Management of eating disorders in primary care Primary care has a particularly important role in the management of those with enduring eating disorders who are not in contact with specialist services.
4.3.2.1 Patients with enduring anorexia nervosa not under the care of a specialist service should be offered an annual physical and mental health review by their GP. (C)
4.4 Anorexia nervosa
4.4.1 Psychological interventions for anorexia nervosa Psychological interventions are a key element in the management of anorexia. There is evidence for the effectiveness of a number of interventions and patient preference will be a key factor in choosing the most appropriate one. For children and adolescents the involvement of family and carers will be important whatever the nature of the intervention. The delivery of psychological interventions should be accompanied by regular monitoring of
71 CONFIDENTIAL Second draft for consultation September 2003 a patient’s physical state including weight and specific indicators of increased physical risk.
Common elements of the psychological treatment of anorexia nervosa
4.4.1.1 Therapies to be considered for the psychological treatment of anorexia nervosa include cognitive analytical therapy (CAT), cognitive behaviour therapy (CBT), interpersonal psychotherapy (IPT), focal psychodynamic therapy and family interventions focused explicitly on eating disorders. (C)
4.4.1.2 Patient and, where appropriate, carer preference should be taken into account in deciding which psychological treatment is to be offered. (C)
4.4.1.3 The aims of psychological treatment should be to reduce risk, encourage weight gain, healthy eating, and reduce other eating- disorder related-symptoms and to facilitate psychological and physical recovery. (C)
Outpatient psychological treatments for adults
4.4.1.4 Most adults with anorexia nervosa should be managed on an outpatient basis with psychological treatment (with physical monitoring) provided by a healthcare professional competent in the psychological treatment and assessment of physical risk of people with eating disorders. (C)
4.4.1.5 Outpatient psychological treatment and physical monitoring for anorexia nervosa should normally be of at least 6 months’ duration. (C)
4.4.1.6 For adults with anorexia nervosa, if during outpatient psychological treatment there is significant deterioration, or the completion of an adequate course of outpatient psychological treatment does not lead to any significant improvement, more intensive forms of treatment (for example a move from individual therapy to combined individual and family work; or day-care, or inpatient care) should be considered. (C)
4.4.1.7 Dietary counselling should not be provided as the sole treatment for anorexia nervosa. (C)
Psychological aspects of inpatient care
72 CONFIDENTIAL Second draft for consultation September 2003
Psychological treatment is often a key element of an inpatient stay but evidence for what kind of treatment or approaches to treatment are effective is limited.
4.4.1.8 For inpatients with anorexia nervosa, a structured symptom-focussed treatment regimen with the expectation of weight gain should be provided in order to achieve weight restoration. It is important to carefully monitor the patient’s physical status during refeeding. (C)
4.4.1.9 Psychological treatment should be provided which has a focus both on eating behaviour and attitudes to weight and shape, and wider psychosocial issues with the expectation of weight gain. (C)
4.4.1.10 Rigid inpatient behaviour modification programmes should not be used in the management of anorexia nervosa. (C)
Post-hospitalisation psychological treatment for people with anorexia nervosa
For patients with anorexia nervosa following discharge from hospital it is usually necessary to extend the duration of psychological treatment over that normally provided to those who have not been hospitalised.
4.4.1.11 Following inpatient weight restoration people with anorexia nervosa should be offered outpatient psychological treatment that focuses both on eating behaviour and attitudes to weight and shape, and wider psychosocial issues with regular monitoring of both physical and psychological risk. (C)
4.4.1.12 The length of outpatient psychological treatment and physical monitoring following inpatient weight restoration should typically be of at least 12 months’ duration. (C)
Children and adolescents with anorexia nervosa
Special considerations are needed in the treatment of children and adolescents. The involvement of families and other carers is particularly important here.
4.4.1.13 Family interventions focused explicitly on eating disorders should be offered to children and adolescents with anorexia nervosa. (B)
4.4.1.14 Children and adolescents with anorexia nervosa should be offered individual appointments with a healthcare professional separate from those with their family members or carers. (C)
73 CONFIDENTIAL Second draft for consultation September 2003
4.4.1.15 The therapeutic involvement of siblings and other family members should be considered in all cases because of the effects of anorexia nervosa on other family members. (C)
4.4.1.16 In children and adolescents with anorexia nervosa the need for inpatient treatment and the need for urgent weight restoration should be balanced alongside the educational and social needs of the young person. (C)
4.4.2 Pharmacological interventions for anorexia nervosa There is a very limited evidence base for the pharmacological treatment of anorexia nervosa. A range of drugs may be used in the treatment of comorbid conditions but caution should be exercised in their use given the physical vulnerability of many people with anorexia nervosa.
4.4.2.1 Medication should not be used as the sole or primary treatment for anorexia nervosa. (C)
4.4.2.2 Caution should be exercised in the use of medication for comorbid conditions such as depression or obsessive-compulsive disorder, as many of these features may resolve with weight gain alone. (C)
4.4.2.3 When medication is used to treat people with anorexia nervosa, the side effects of drug treatment, (in particular, cardiac side effects), should be carefully considered because of the compromised cardiovascular function of many people with anorexia nervosa. (C)
4.4.2.4 Healthcare professionals should be aware of the risk of drugs which prolong the QTc interval; for example, antipsychotics, tricyclic antidepressants, macrolide antibiotics, and some antihistamines. (C)
4.4.2.5 In patients with anorexia nervosa with borderline/ prolonged QTc interval the prescription of drugs with side effects that may compromise cardiac functioning should be avoided. (C)
4.4.2.6 If the prescription of medication that may compromise cardiac functioning is essential, ECG monitoring should be undertaken. (C)
4.4.2.7 All patients with a diagnosis of anorexia nervosa should have an alert placed in their prescribing record concerning the risk of side effects. (C)
4.4.3 Physical management of anorexia nervosa Anorexia nervosa carries considerable risk of serious physical morbidity. Awareness of the risk, careful monitoring and, where appropriate, close
74 CONFIDENTIAL Second draft for consultation September 2003
liaison with an experienced physician are important in the management of the physical complications of anorexia nervosa.
Managing weight gain
4.4.3.1 In most patients with anorexia nervosa an average weekly weight gain of 0.5-1 kg in inpatient settings and 0.5 kg in outpatient settings should be an aim of treatment. This requires about 3,500 to 7000 extra calories a week. (C)
4.4.3.2 Healthcare professionals should ensure for children and adolescents with anorexia nervosa that once a healthy weight is reached the increased energy and necessary nutrients are available in the diet to support growth and development. (C)
4.4.3.3 Regular physical monitoring and in some cases treatment with multi- vitamin/ multi-mineral supplement in oral form is recommended for people with anorexia nervosa during both inpatient and outpatient weight restoration. (C)
4.4.3.4 In the nutritional management of children and adolescents with anorexia nervosa, carers need to be included in any dietary education or meal planning. (C)
4.4.3.5 Total parenteral nutrition should not be used for people with anorexia nervosa, unless there is a specific indication such as gastrointestinal dysfunction. (C)
Managing risk
4.4.3.6 Healthcare professionals should monitor physical risk in patients with anorexia nervosa and where this leads to the identification of increased physical risk, the frequency and the monitoring and nature of the investigations should be adjusted accordingly. (C)
4.4.3.7 People with anorexia nervosa and their carers should be informed if the risk to their physical health is high. (C)
4.4.3.8 The involvement of a physician or paediatrician with expertise in the treatment of physically at-risk patients with anorexia nervosa should be considered for all individuals who are physically at-risk. (C)
4.4.3.9 Pregnant women with either current or remitted anorexia nervosa may need more intensive prenatal care to ensure adequate prenatal nutrition and fetal development. (C)
75 CONFIDENTIAL Second draft for consultation September 2003
4.4.3.10 Oestrogen administration should not be used to treatment bone density problems in children and adolescents as this may lead to premature fusion of the epiphyses. (C)
Feeding against the will of the patient
4.4.3.11 Feeding against the will of the patient should be an intervention of last resort in the care and management of anorexia nervosa. (C)
4.4.3.12 Feeding against the will of the patient is a highly specialised procedure requiring expertise in the care and management of those with severe eating disorders and the physical complications associated with it. This should only be done in the context of the Mental Health Act 1983 or Children Act 1989. (C)
4.4.3.13 When making the decision to feed against the will of the patient the legal basis for any such action must be clear. (C)
4.4.4 Service interventions The following section considers those aspects of the service system relevant to the treatment and management of anorexia nervosa. It pays particular attention to the use of inpatient services and the particular challenges in the inpatient management of children and adolescents with anorexia nervosa.
4.4.4.1 Inpatient treatment should be considered for people with anorexia nervosa whose disorder is associated with high or moderate physical risk. (C)
4.4.4.2 Patients with anorexia nervosa requiring inpatient treatment should normally be admitted to a setting that can provide the skilled implementation of refeeding with careful physical monitoring and in combination with psychosocial interventions. (C)
4.4.4.3 Inpatient treatment or day patient treatment should be considered for people with anorexia nervosa whose disorder has not improved with appropriate outpatient treatment, or for whom there is a significant risk of suicide or severe self-harm. (C)
4.4.4.4 Healthcare professionals without specialist experience, or in situations of uncertainty, should consider seeking advice from an appropriate specialist when considering a formal admission for a patient with anorexia nervosa regardless of the age of the patient. (C)
4.4.4.5 Healthcare professionals managing patients with anorexia nervosa, especially that of the binge purging sub-type, should be aware of the increased risk of self-harm and suicide, particularly at times of transition between services or service settings. (C)
76 CONFIDENTIAL Second draft for consultation September 2003
4.4.5 Special considerations for children and adolescents
4.4.5.1 Admission of children and adolescents with anorexia nervosa should be to age-appropriate facilities (with the potential for separate children and adolescent services), which have the capacity to provide appropriate educational and related activities. (C)
4.4.5.2 When a young person with anorexia nervosa refuses treatment, which is deemed essential, consideration should be given to the use of the Mental Health Act 1983 or the right of those with parental responsibility to override the young person’s refusal. (C)
4.4.5.3 Continuing treatment relying indefinitely on parental consent should be avoided. It is recommended that the legal basis under which treatment is being carried out should be recorded in the patient’s case notes in all cases but particularly in the case of children and adolescents. (C)
4.4.5.4 For children and adolescents’ with anorexia nervosa, where issues of consent to treatment are highlighted, healthcare professionals should consider seeking a second opinion. (C)
4.4.5.5 If the patient with anorexia nervosa and those with parental responsibility refuse treatment, and treatment is deemed to be essential, legal advice should be sought in order to consider proceedings under the Children Act 1989. (C)
4.5 Bulimia nervosa
4.5.1 Psychological interventions Psychological interventions are the key element in the treatment and management of bulimia nervosa. There is evidence for the effectiveness of specific psychological interventions and some evidence for the use of self-help as part of a stepped-care approach.
4.5.1.1 As a possible first step, patients with bulimia nervosa should be encouraged to follow an evidence-based self-help programme. (B)
4.5.1.2 Healthcare professionals should consider providing direct encouragement and support to patients undertaking an evidence- based self-help programme as this may improve outcomes. This may be sufficient treatment for a limited subset of patients. (B)
4.5.1.3 Cognitive behaviour therapy for bulimia nervosa (CBT-BN), a specifically adapted form of cognitive behaviour therapy, should be offered to adults with bulimia nervosa. (A)
77 CONFIDENTIAL Second draft for consultation September 2003
4.5.1.4 An adequate course of CBT-BN should normally be of 16 to 20 sessions over 4 to 5 months. (A)
4.5.1.5 Older adolescents with bulimia nervosa may be treated with CBT-BN adapted as needed to suit their age, circumstances and level of development. (C)
4.5.1.6 When people with bulimia nervosa have not responded to or do not want CBT, other psychological treatments should be considered. (B)
4.5.1.7 Interpersonal psychotherapy should be considered as an alternative to cognitive behaviour therapy, but patients should be informed it takes 8 - 12 months to achieve results comparable to cognitive behaviour therapy. (B)
4.5.2 Pharmacological interventions There is good evidence that pharmacological treatment can benefit patients with bulimia nervosa in the short term but very little is known about the maintenance of these beneficial effects. Antidepressants may be used alone or in combination with psychological treatment to increase the likelihood of a response to treatment.
4.5.2.1 As an alternative or additional first step to using an evidence-based self-help programme, adults with bulimia nervosa may be offered a trial of an antidepressant drug. (B)
4.5.2.2 Patients should be informed that antidepressant drugs can reduce the frequency of binge eating and purging, but the long-term effects are unknown. Any beneficial effects will be rapidly apparent. (B)
4.5.2.3 Selective serotonin reuptake inhibitors (SSRIs) (specifically fluoxetine) are the drugs of first choice for the treatment of bulimia nervosa in terms acceptability, tolerability and reduction of symptoms. (C)
4.5.2.4 For people with bulimia nervosa, the effective dose of fluoxetine is higher than for depression (60 mg daily). (C)
4.5.2.5 For people with bulimia nervosa who are not responding to CBT-BN, the addition of an antidepressant drug should be considered as this may improve outcome. (B)
4.5.2.6 No drugs, other than antidepressants, are recommended for the treatment of bulimia nervosa. (B)
78 CONFIDENTIAL Second draft for consultation September 2003
4.5.3 Management of physical aspects of bulimia nervosa Patients with bulimia nervosa can experience considerable physical problems as a result of a range of behaviours associated with bulimia nervosa. Awareness of the risks, and careful monitoring should be a concern of all healthcare professionals working with people with this disorder.
4.5.3.1 Patients with bulimia nervosa who are vomiting frequently or taking large quantities of laxatives (and especially if they are also underweight) should have their fluid and electrolyte balance assessed. (C)
4.5.3.2 When electrolyte disturbance is detected, it is usually sufficient to focus on eliminating the behaviour responsible. In the small proportion of cases where supplementation is required to restore the patient’s electrolyte balance oral rather than intravenous administration is recommended, unless there are problems with gastro intestinal absorption. (C)
4.5.3.3 Where laxative abuse is present, patients should be advised to gradually reduce laxative use and informed that laxative use does not significantly reduce calorie absorption. (C)
4.5.3.4 Treatment of both subthreshold and clinical cases of bulimia nervosa in people with diabetes is essential because of the greatly increased physical risk in this group. (C)
4.5.3.5 Patients with Type 1 Diabetes and bulimia nervosa should have intensive regular physical monitoring as they are at high risk of retinopathy and other complications. (C)
4.5.3.6 Pregnant women with bulimia nervosa require careful monitoring throughout the pregnancy and in the post-partum period. (C)
4.5.4 Dental care It should be noted that patients with an eating disorder other than bulimia nervosa are also at risk of poor dental health if vomiting is a symptom of the disorder.
4.5.4.1 Patients with bulimia nervosa should have regular dental reviews. (C)
4.5.4.2 Patients who are vomiting should be given appropriate advice on dental hygiene including avoiding brushing after vomiting, rinsing with a non-acid mouth wash after vomiting, and measures to reduce an acid oral environment (e.g., limiting acid foods). (C)
79 CONFIDENTIAL Second draft for consultation September 2003
4.5.5 Service interventions The great majority of patients with bulimia nervosa can be treated as outpatients. There is a very limited role for the inpatient treatment of bulimia nervosa. This is primarily concerned with the management of suicide risk or severe self-harm.
4.5.5.1 The great majority of patients with bulimia nervosa should be treated in an outpatient setting. (C)
4.5.5.2 For patients with bulimia nervosa who are at risk of suicide or severe self-harm, admission as an inpatient or a day patient or the provision of more intensive outpatient care, should be considered. (C)
4.5.5.3 Psychiatric admission for people with bulimia nervosa should normally be undertaken in a setting with experience of managing this disorder. (C)
4.5.5.4 Healthcare professionals should be aware that patients with bulimia nervosa who have poor impulse control, notably substance misuse, may be less likely to respond to a standard programme of treatment. As a consequence treatment should be adapted to the problems presented. (C)
4.6 Treatment and management of atypical eating disorders including binge eating disorder Binge eating disorder (BED), has a growing evidence base relating to psychological treatment and self-help approaches. Little is known about the effective treatment of the comorbid obesity, which often accompanies binge eating disorder.
4.6.1 Psychological treatments for binge eating disorder
4.6.1.1 As a possible first step, patients with binge eating disorder should be encouraged to follow an evidence-based self-help programme. (B)
4.6.1.2 Healthcare professionals should consider providing direct encouragement and support to patients undertaking an evidence- based self-help programme as this may improve outcomes. This may be sufficient treatment for a limited subset of patients. (B)
4.6.1.3 Cognitive behaviour therapy for binge eating disorder (CBT-BED), a specifically adapted form of CBT, should be offered to adults with binge eating disorder. (A)
80 CONFIDENTIAL Second draft for consultation September 2003
4.6.1.4 Other psychological treatments (interpersonal psychotherapy for binge eating disorder, and modified dialectical behaviour therapy) may be offered to adults with persistent binge eating disorder. (B)
4.6.1.5 Patients should be informed that all psychological treatments for binge eating disorder have a limited effect on body weight. (A)
4.6.1.6 When providing psychological treatments for patients with binge eating disorder consideration should be given to the provision of concurrent or consecutive interventions focusing on the management of comorbid obesity. (C)
4.6.1.7 Suitably adapted psychological treatments should be offered to adolescents with persistent binge eating disorder. (C)
4.6.2 Pharmacological interventions for binge eating disorder
4.6.2.1 As an alternative or additional first step to using an evidence-based self-help programme or a behavioural weight control programme, consideration should be given to offering a trial of a SSRI antidepressant drug to patients with binge eating disorder. (B)
4.6.2.2 Patients with binge eating disorders should be informed that SSRIs can reduce binge eating, but the long-term effects are unknown. Antidepressant drugs may be sufficient treatment for a limited subset of patients. (B)
4.6.3 General treatment of atypical eating disorders
4.6.3.1 In the absence of evidence to guide the management of atypical eating disorders (eating disorders not otherwise specified) other than binge eating disorder, it is recommended that the clinician considers following the guidance on the treatment of the eating problem that most closely resembles the individual patient’s eating disorder. (C)
4.7 Research recommendations The GDG identified the following research recommendations specifically to address gaps in the evidence base: • Adequately powered efficacy studies of specific treatments and services for people with anorexia nervosa are required
• Adequately powered efficacy studies of the treatment for of atypical eating disorders (eating disorders not otherwise specified) are required
81 CONFIDENTIAL Second draft for consultation September 2003
• Adequately powered efficacy studies of the treatment of adolescents with bulimia nervosa and non-responders to cognitive behaviour therapy are required
• Effectiveness studies of the treatment of bulimia nervosa in adults are required
• Patient and carer satisfaction is an important outcome and may influence treatment approaches. It should be considered a routine outcome in research
• Further research is needed to assess the validity of screening instruments in primary care.
82 CONFIDENTIAL Second draft for consultation September 2003
5 Identification of eating disorders in primary care
5.1 Introduction Although new cases of clinical anorexia nervosa are not a common occurrence in primary care (an average GP list of 1900 will have only 1 or 2 sufferers), eating disorders including bulimia nervosa and EDNOS reach a prevalence of 5% in young women. Community studies show that less than half of clinical cases of eating disorders are identified in primary care. Despite this patients with EDs consult more frequently prior to diagnosis with a variety of symptoms, psychological, gynaecological and gastroenterological (Ogg et al., 1997). The difficulties facing primary care clinicians in diagnosis derive in part from illness related factors, such as ambivalence, denial, secrecy and shame, which make it difficult for sufferers to be open with their doctors. GPs may have little experience with EDs, and feel anxious about their management or unsympathetic towards an illness that may appear in part self inflicted. Patients may report that the problem is not always taken seriously enough if presented at an early stage. Practitioners risk failing to identify EDs if they do not consider the fact that these illnesses also occur in groups not traditionally considered to be at risk, such as, children, men and those from ethnic minority groups, and lower social classes.
The effective management of anorexia nervosa depends on a full assessment of physical status, psychological features, risk and capacity to consent to treatment. The details of appropriate physical assessment are detailed in section 5.2.4, but included in the assessment of high physical risk will be an assessment of degree of emaciation (BMI/BMI centile), the presence of purging and any fluid restriction. The rate of weight loss is also a helpful indicator. In children and younger adolescents, reference to BMI norms is necessary, but it is also of note that children’s relatively small fat stores render them at risk with relatively less weight loss. Psychosocial assessment should take account of motivation and social support.
5.1.1 Current practice Little is known about how EDs are actually assessed or managed in primary care. Referrals to specialist centres vary widely as does the information contained in them. Current practice is not underpinned by an evidence base derived from primary care settings and guidance has largely been extrapolated from secondary or tertiary care settings with experience with more severely ill clinical populations.
83 CONFIDENTIAL Second draft for consultation September 2003
5.2 Screening Early detection and treatment of eating disorders may improve outcomes in eating disorders. General practitioners and other members of the primary care team are in a good position to identify patients with eating problems early. Screening tools may facilitate this process. The most effective screening device probably remains the general practitioner thinking about the possibility of an eating disorder.
It would be impractical for general practitioners to try and screen all their patients for eating disorders, as the prevalence of eating disorders in the general population is low. It might be possible to screen new patients when they register. One or two screening questions could be used to raise the index of suspicion, either verbally during the registration health check or in writing as part of the registration questionnaire.
High risk groups within the general practice population could be targeted opportunistically. Such groups include young women, patients with a low or high BMI, adolescents consulting with weight concerns, menstrual disturbances or amenorrhoea, gastrointestinal disorders and psychological problems. A brief screening questionnaire could be used for such high-risk groups.
5.2.1 Current practice Eating disorders may be difficult to detect in primary care settings. Patients may be slow to self-present and many remain undetected by general practitioners (King, 1989; Whitehouse et al, 1992). Adults with eating disorders appear to consult their general practitioner more frequently than controls, presenting particularly with psychological, gastrointestinal and gynaecological problems (Ogg et al., 1997). Consultations of this nature present an opportunity to screen for eating disorders. At present, no formal screening tool for eating disorders is widely used in primary care.
5.2.2 Screening for eating disorders The aim of screening is to facilitate detection so that treatment can be offered early in the course of the eating disorder.
A systematic review of the literature did not identify a significant body of work in this area, nor were any high quality existing systematic reviews identified. The absence of high quality evidence in this area inevitably limits the conclusions that can be drawn from the review. The relevant studies that were identified are described below.
Existing screening instruments A range of questionnaires exists of which the Eating Attitudes Test, EAT (Garner & Garfinkel, 1979) is probably the most widely used as a screening tool in epidemiological studies. In addition there are a number of other pencil
84 CONFIDENTIAL Second draft for consultation September 2003
and paper measures to assess eating disorder psychopathology (e.g., the Eating Disorder Inventory, EDI - Garner et al., 1983). However, these take a long time to administer and may need to be interpreted by specialists. Such instruments may be well suited for evaluating treatment progress in patients with eating disorders, but may not perform well in screening for eating disorders in community samples due to symptom denial and low prevalence (Williams et al., 1982; Carter &Moss, 1984).
Questionnaires of this type may have a role for screening in very high risk groups in special settings, e.g., in ballet schools, fitness and sports facilities. They may have occasional application in general practice, when a patient with a probable eating disorder has already been identified.
Several brief screening questionnaires, more suitable for use in community samples, have been developed and evaluated. These include the SCOFF (Morgan et al., 1999), Anstine and Grinenko (2000), the BITE and the BES (Ricca et al., 2000), the EDS-5 (Rosenvinge, 2001), Freund et al., 1993, the ESP (Cotton et al., 2003), Ri-BED-8 (Waaddegaard et al., 1999, the EDDS (Stice et al., 2000), the EAT-12 and the EDE-S/Q (Beglin & Fairburn, 1992, 1994). The most promising to date is the SCOFF.
The SCOFF questionnaire (Morgan et al., 1999; Luck et al., 2002; Perry et al., 2002) was developed and validated in the UK. It consists of five questions designed to clarify suspicion that an eating disorder might exist rather than to make a diagnosis. The questions can be delivered either verbally or in written form and there is one study validating the use of the SCOFF in adult women in a general practice population (Luck, 2002). Further research is needed to evaluate the SCOFF questions in general practice populations before they can be recommended for use in primary care.
5.2.3 Clinical summary A number of brief screening methods have been developed that have some utility in detecting eating disorders. The SCOFF has been shown to be capable of determining cases of eating disorders in adult women in primary care. The place of longer questionnaires (e.g., EAT, EDI, BITE, EDE-Q) may be in further assessment once index of suspicion has been raised. They may also be useful to facilitate decisions regarding referral to specialist services. Certain clinical presentations should also raise the index suspicion, for example, adolescent girls with concerns about weight, and women consulting with menstrual disturbances, gastrointestinal or psychological symptoms.
5.2.4 Identification The most important factor in the identification of eating disorders in generalist settings is for the practitioner to consider the possibility of an eating disorder and to be prepared to inquire further in an empathic and non- judgmental manner. The history is paramount and special investigations are not normally required to make a diagnosis
85 CONFIDENTIAL Second draft for consultation September 2003
5.2.4.1 Anorexia nervosa The first contact with healthcare services is often made by a worried family member, friend or schoolteacher rather than the patient. Concerns expressed may be related to weight loss, food related behaviours such as skipping meals, hiding food or adopting a restrictive diet. There may be a change in mood, sleep patterns and increased activity. Typical psychopathological features are fear of gaining weight or becoming fat despite being underweight, disturbance in evaluating or experiencing body weight or shape, undue influence of eating or changes in body weight on self-evaluation and preoccupation with shape or weight related matters. These features may not all be present, easy to elicit or they may be denied. However, denial of the seriousness of the weight loss or consequences, both physical and psychological is usually present. Established anorexia nervosa with signs of emaciation is usually obvious. However, patients may present initially in primary care with non-specific physical symptoms such as abdominal pain, bloating, constipation, cold intolerance, light headedness, hair, nail or skin changes. Amenorrhoea, combined with unexplained weight loss, in the population at risk should always prompt further enquiry. Apparent food allergy/intolerance and chronic fatigue syndrome sometimes precede the development of an eating disorder and may cause diagnostic confusion. In children, growth failure may be a presenting feature.
In practice, typical cases should cause little difficulty when the time is taken to explore the history including corroborative information and the patient’s attitude to the weight loss. Indeed, diagnosis is often delayed when doctors inadvertently collude by over-investigating and referring to other specialties rather than confronting the possibility of an eating disorder.
Diagnostic criteria These are outlined in the introduction (See section 2.1.2).
In children and adolescents under 18, the use of BMI centile charts should be encouraged (Cole, Bellizi et al., 2000) with a cut off less than 10th centile of the reference population indicating underweight.
Centile charts for weight and height are also helpful in showing failure to progress over time.
The list of potential differential diagnoses of weight loss or amenorrhoea is large, but in practice, typical cases should cause little difficulty when the time is taken to explore the history including corroborative information and the patient’s attitude to the weight loss. The following factors need to be considered:
86 CONFIDENTIAL Second draft for consultation September 2003
• Risk factors - family history of eating disorder, Type 1 Diabetes, previously overweight, occupation (e.g., athlete, dancer, model). Although adolescent girls and young women constitute the principal population at risk, it should be remembered that eating disorders also occur in ethnic minorities, men and children
• Differential diagnosis of weight loss - includes malabsorbtion (e.g., coeliac disease inflammatory bowel disease), neoplasm, illicit drug use, infection (e.g., TB), autoimmune disease, endocrine disorders (e.g., hyperthyroidism)
• Differential diagnosis of amenorrhoea – includes pregnancy, primary ovarian failure, poly cystic ovary syndrome, pituitary prolactinoma, uterine problems and other hypothalmic causes
• Psychiatric differential diagnosis includes – depression, obsessive compulsive disorder, somatisation and, rarely, psychosis.
Initial physical assessment The rationale for physical assessment is more to determine the presence and severity of emaciation and secondary physical consequences of the anorexia nervosa than to ascertain the primary diagnosis.
It should include as a minimum:
• Height weight and BMI
• Centile charts for age less than 18
• Pulse and blood pressure.
The following may also be helpful to assess the risk of physical instability:
• Core temperature (this is easily done by ear thermometer)
• Examination of peripheries (Circulation and oedema)
• Cardiovascular examination including postural hypotension
• Situp/squat test (A test of muscle power) (Robinson, 2003).
Laboratory investigations
87 CONFIDENTIAL Second draft for consultation September 2003
Extensive laboratory investigation is not usually required in the diagnosis or assessment of anorexia nervosa in a primary care setting. Many tests remain normal even with extreme weight loss and are a poor guide to physical risk. The diagnosis is always made on the basis of the clinical history.
Investigations The following would represent a reasonable initial screen in primary care if there are no other indications or diagnostic concerns:
• Full Blood Count, ESR, Urea & Electrolytes, Creatinine, Liver Function Tests, Random Blood Glucose, Urinalysis
• ECG: This should be considered in all cases and is essential if symptoms/signs of cardiac compromise, bradycardia, electrolyte abnormality or BMI less than 15 kg/m2. (Or equivalent on centile chart).
Further tests may be required in more severe cases or to assess complications: Calcium, Magnesium, Phosphate, Serum Proteins, Creatine Kinase (CK or CPK).
Tests that may be needed in the differential diagnosis of amenorrhoea and weight loss:
• Thyroid Function Tests, Follicle Stimulating Hormone, Lutenising Hormone, Prolactin, Chest X-Ray.
A DEXA scan may be considered for identification of osteopenia/ osteoporosis, which may occur after 6-12 months of amenorrhoea. Although this is not necessarily a primary care level investigation, it may be helpful in encouraging motivation for change in those not yet ready to accept referral, by demonstrating the real physical consequences of the anorexia.
5.2.4.2 Bulimia nervosa
Identification The patient with bulimia nervosa is more likely to be older and to consult alone than a patient with anorexia nervosa. There may be a history of previous anorexia nervosa or of unhappiness with previous weight and attempts to diet. Appropriate questioning (see screening section above) may reveal patterns of restriction, binge eating and purging and psychopathology that make the diagnosis clear. Not infrequently, physical symptoms are presented which may be related to or consequences of purging or laxative
88 CONFIDENTIAL Second draft for consultation September 2003
use. These symptoms, particularly in a young woman should be a “red flag” in prompting the practitioner to consider further enquiry.
Where the patient does not disclose bulimia nervosa, a range of symptoms may present which should raise the index of suspicion. These include requests for help with weight loss, menstrual disturbance and the physical consequences of vomiting and laxative and diuretic use. Non-specific symptoms may include fatigue, lethargy. Gastrointestinal disorders may be present including bloating, fullness, abdominal pain, irritable bowel syndrome type symptoms, constipation, diarrhoea and rectal prolapse as well as oesophagitis and gastrointestinal bleeding. Oropharyngeal symptoms may include a sore throat, parotid swelling and dental enamel erosion.
Physical examination and investigation In bulimia nervosa and related conditions, characteristic physical signs have been described (e.g., parotid enlargement, Russell’s sign (callus formation on the dorsum of the hand) and dental enamel erosion, which are usually manifestations of purging. In practice these are not seen in the majority of patients presenting in primary care with bulimic disorders, although electrolyte abnormalities are reasonably common, so urea and electrolytes should be routinely obtained. These are covered in Section 7.5.2.
5.2.5 Clinical practice recommendations
5.2.5.1 For people with eating disorders presenting in primary care, GPs should take responsibility for the initial assessment and the initial coordination of care. This includes the determination of the need for emergency medical or psychiatric assessment. (C)
5.2.5.2 Where management is shared between primary and secondary care, there should be clear agreement between individual healthcare professionals on the responsibility for monitoring of patients with eating disorders. This agreement should be in writing and should be shared with the patient and, where appropriate, families and carers. (C)
5.2.5.3 Target groups for screening questions should include young women with low body mass index (BM) compared with age norms, patients consulting with weight concerns who are not overweight, menstrual disturbances or amenorrhoea, gastrointestinal symptoms, patients with physical signs of starvation or repeated vomiting and children with poor growth. (C)
5.2.5.4 When screening for eating disorders one or two simple questions should be considered for use with specific target groups (for example, “Do you think you have an eating problem?” and ”Do you worry excessively about your weight?”). (C)
89 CONFIDENTIAL Second draft for consultation September 2003
5.2.5.5 )
5.2.5.6 Young women with Type 1 Diabetes and poor treatment adherence should be screened and assessed for the presence of an eating disorder. (C)
5.2.5.7 In anorexia nervosa, although weight and body mass index (BMI) are important indicators of physical risk they should not be considered the sole indicators (as on their own they are unreliable in adults and especially in children). (C)
5.2.5.8 In assessing whether a person has anorexia nervosa, attention should be paid to the overall clinical assessment (repeated over time), including rate of weight loss, growth rates in children, objective physical signs and appropriate laboratory tests. (C)
5.2.5.9 Patients with enduring anorexia nervosa not under the care of a specialist service should be offered an annual physical and mental health review by their general practitioner. (C)
90 CONFIDENTIAL Second draft for consultation September 2003
6 Treatment and management of anorexia nervosa
6.1 Introduction The treatment plan for a patient with anorexia nervosa needs to consider the appropriate service setting, and the psychological and physical management, but unfortunately the research evidence base to guide decision-making is very limited. The appropriate setting depends on the assessment of risk and the patient’s wishes, but in general the person with anorexia nervosa will initially be treated in a secondary care outpatient service, moving into a day or inpatient setting if required. Although convincing evidence is lacking on the most effective form of psychological therapy, psychological therapy is nevertheless crucial in addressing the underlying behaviours and cognitions. In children and adolescents some family based psychological intervention is essential. Physical treatments comprise nutritional interventions and psychopharmacological agents. The latter are used to support psychological treatments or for the management of co-morbid conditions, rather than being first line treatments.
The treatment options should be discussed fully with the patient in order that he or she can make informed choices. Given the ambivalence inherent in this disorder, engagement and efforts at motivational enhancement may be helpful in maximising adherence to treatment. A small number of patients with anorexia nervosa do not have the capacity to make decisions about their own health and safety and in these cases provision for their admission to hospital and treatment is under the remit of the Mental Health Act 1983 and the Children Act 1989. 6.2 Psychological interventions
6.2.1 Introduction The earliest models of psychological treatment for anorexia nervosa were psychodynamic in nature, albeit with few attempts to study them systematically (for review see Dare & Crowther, 1995; Herzog & Hartmann, 1997; Kaplan, 2002).
From the 1960s onwards, behaviour therapy for anorexia nervosa became increasingly popular, with many articles reporting on the use of operant conditioning techniques (for review see Bemis et al., 1987; Fairburn & Cooper, 1987; Schmidt, 1989). The basic operant paradigm consisted of isolating patients from social and material reinforcers that were subsequently delivered
91 CONFIDENTIAL Second draft for consultation September 2003
contingent on weight gain or caloric intake. For a time, this technique became part of many inpatient regimes, because of its efficacy in encouraging weight gain in the short-term. However, operant conditioning approaches have been criticised as coercive and controlling (for review see Bemis, 1987), and their influence has waned in recent years.
The seminal – though controversial – work of Minuchin and colleagues (1975) stimulated much interest in the use of family interventions in the treatment of anorexia nervosa. Initially, the rationale for this approach was rooted in the notion that families have a key causal role in the development of anorexia nervosa, “the anorexogenic family”. However, it is now widely agreed that family interventions are best viewed as treatments that mobilise family resources rather than treating family dysfunction, for which there is no empirical evidence (Eisler et al., 2003). The first treatment trial of family therapy was published in 1987 (Russell et al., 1987), studying patients who had undergone a period of weight restoration in a specialist inpatient unit prior to starting outpatient psychotherapy. This study showed that in 21 adolescents with a short duration of illness, family therapy was superior to individual supportive counselling in maintaining weight gained. The findings of this study stimulated three further RCTs into different types of family interventions for adolescents with anorexia nervosa (Le Grange et al., 1992; Eisler et al., 2000; Geist et al., 2002). In addition there has been one further comparison of family therapy with individual therapy although the findings are difficult to interpret (Robin et al., 1999). The original Maudsley model of family therapy has since been manualised for therapists (Lock et al., 2001).
A handful of controlled trials, mainly in adults with anorexia nervosa, have evaluated the efficacy of specific individual psychotherapies, such as cognitive behaviour therapy (Channon et al., 1989, Serfaty et al., 2002, Pike et al., in press, MacIntosh et al., submitted); cognitive analytical psychotherapy (Treasure et al., 1995; Dare et al., 2001), focal analytical psychotherapy (Dare et al., 2001) and interpersonal psychotherapy (McIntosh et al., submitted).
More recently there has been interest in the use of motivational interventions in the engagement and treatment of people with anorexia nervosa (Treasure & Ward, 1997; Vitousek et al., 1998), however, RCTs in this area are as yet lacking.
Overall, the body of research into the treatment of anorexia nervosa is small and inconsistent in methodological quality. The conclusions that can be drawn are limited because many studies have no follow-up data, lack the statistical power necessary to detect real effects, and use different study entry criteria and outcome measures.
92 CONFIDENTIAL Second draft for consultation September 2003
6.2.2 Current practice There is wide variability in the availability of psychological therapies for patients with anorexia nervosa. There is no uniform or agreed approach to the psychological treatment or management of anorexia nervosa in adults, either in terms of types of treatment offered, their duration, intensity or the setting in which treatment is provided.
In the treatment of anorexia nervosa in children and adolescents, family interventions are usually offered. These may vary in approach and not all of them will resemble the evidence-based family interventions that have a focus on eating behaviours. Specialist services may offer a range of individual psychological therapies including cognitive behaviour therapy, psychodynamic psychotherapy, motivational enhancement therapy and family interventions.
Principles of psychological treatment: engagement Many people with anorexia nervosa find it hard to acknowledge that they have a problem and are ambivalent about change. This contributes to their reluctance to engage with treatment and services. A precondition for any successful psychological treatment is the effective engagement of the patient in the treatment plan. Healthcare professionals involved in the treatment of anorexia nervosa should take time to build an empathic, supportive and collaborative relationship with patients and, if applicable, their carers. This should be regarded as an essential element of the care offered. Motivation to change may go up and down over the course of treatment and the therapist needs to remain sensitive to this. Special challenges in the treatment of anorexia nervosa include the highly positive value placed by people with anorexia nervosa on some of their symptoms, and their denial of the potentially life-threatening nature of their disorder.
Aims of psychological interventions In general, the aims of psychological treatment are to promote weight gain and healthy eating, to reduce other eating disorder related symptoms and to promote psychological recovery. In patients who have just had their weight restored in hospital the maintenance of weight gain is a prominent goal, together with continued healthy eating, the reduction of other eating disorder related symptoms and the promotion of psychological recovery. In patients with enduring anorexia nervosa, psychological treatment may have more modest goals and may focus on improving quality of life and maintaining a stable or safe weight rather than aiming for an optimal weight.
6.2.3 Outpatient psychological treatments (first episode and later episodes) This section focuses on psychological treatments given as the main or only treatment to patients who present during a first or later episode of anorexia nervosa.
93 CONFIDENTIAL Second draft for consultation September 2003
6.2.3.1 Psychological treatments reviewed The following treatments were included: • Behaviour therapy (BT)
• Cognitive-analytical therapy (CAT)
• Cognitive behaviour therapy (CBT)
• Interpersonal psychotherapy (IPT)
• Family therapy and family interventions
• Psychodynamic psychotherapy
• Psychological treatment not otherwise specified (Psychotherapy NOS).
The Psychological Topic Group established definitions for each treatment (see Glossary). Two members of the Topic Group assessed each study for eligibility and classified each psychological treatment. Where disagreements arose, they were resolved by discussion.
6.2.3.2 Studies considered2 The review team conducted a new systematic search for RCTs of outpatient psychological treatments used for the initiation of treatment during the first or later episodes of anorexia nervosa. Eleven small RCTS (BACHAR1999, CHANNON1989, CRISP1991, DARE2001, EISLER2000, HALL1987, MCINTOSHSUBMIT, ROBIN1999, SERFATY1999, TREASURE1995, WALLIN2000) were identified providing data on 459 participants ranging in age from adolescents to young adults.
Full details of studies included in the guideline and reasons for excluding studies are given in Appendix 17.
6.2.3.3 Evidence statements3 Due to major differences in the way the main outcomes were reported in each study and the lack of extractable data in several studies, no meta-analysis of results was conducted. The statements in this section relate to adults, except where explicitly stated. The level of evidence (I, IIb, IIb, III, IV) is given after
2 Here and elsewhere in the guideline, each study considered for review is referred to by a study ID (primary author and date of study publication, except where a study is in press or only submitted for publication, then a date is not used). 3 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
94 CONFIDENTIAL Second draft for consultation September 2003
each statement (see section 3.4.6 for more information about the classification of evidence).
Effect of outpatient psychological treatments given for the treatment of first episodes or later acute episodes upon symptoms (weight gain and/ or proportion recovered)
There is limited evidence that family interventions, psychotherapy NOS and focal psychoanalytic psychotherapy provided at tertiary referral centres are superior to ’treatment as usual’ in terms of weight gain by end of treatment and post-treatment follow-up (n = 174; CRISP1991, DARE2001). I
There is limited evidence that both family interventions and focal psychoanalytic psychotherapy given at a tertiary referral centre are superior to ’treatment as usual’ in terms of the proportion of people recovered by end of treatment (n = 84; DARE2001). I
There is limited evidence that treatment outcome with different psychological therapies (including BT, CAT, CBT, IPT, family therapy, focal psychodynamic psychotherapy, and psychotherapy NOS) by the end of treatment and at follow-up (up to 5 years) is poor (in terms of weight gain/ proportion of people recovered) in patients referred to tertiary referral centres (n = 258; CHANNON1989, CRISP1991, DARE2001, HALL1987, TREASURE1995). I
There is insufficient evidence to suggest that any particular specialist psychological treatment (including CAT, CBT, IPT, family therapy, focal psychodynamic therapy) is superior to any other in the treatment of adult patients with anorexia nervosa either by the end of treatment or at follow-up (n = 297; BACHAR1999, CHANNON1989, CRISP1991, DARE2001, MCINTOSHSUBMIT, TREASURE1995). I
There is insufficient evidence to determine the efficacy of dietary counselling on its own as a treatment for anorexia nervosa (n = 65; HALL1987, SERFATY1999). I
There is insufficient evidence to determine any advantage for inpatient care over outpatient psychological treatments (individual therapy + family therapy + dietary counselling or group therapy + parents’ group + dietary counselling) for patients who are not so severely ill as to need emergency treatment (n = 90; CRISP1991). I
In children and adolescents, there is insufficient evidence to determine whether conjoint or separate family therapy is more, or less, effective at the end of treatment or follow-up (EISLER2000, ROBIN1999). I
95 CONFIDENTIAL Second draft for consultation September 2003
In children and adolescents, there is insufficient evidence to determine whether the addition of body awareness therapy to family therapy is superior to family therapy alone (WALLIN2000).
Acceptability of outpatient psychological treatment given for the treatment of first episodes or later acute episodes
There is insufficient evidence to determine whether specialist outpatient psychological treatments (including BT, CAT, CBT, family therapy, focal psychodynamic psychotherapy) are more, or less, acceptable to people with anorexia nervosa when compared to ‘standard care’ (n = 198; CHANNON1989, CRISP1991, DARE2001). I
There is insufficient evidence to suggest that any particular specialist psychotherapy (including BT, CAT, CBT, family therapy, focal psychodynamic psychotherapy, IPT, psychotherapy NOS) is more, or less, acceptable to adults with anorexia nervosa (n = 297; BACHAR1999, CHANNON1989, CRISP1991, DARE2001, MCINTOSHSUBMIT, TREASURE1995). I
There is limited evidence to suggest that dietary counselling on its own is less acceptable to people with anorexia nervosa when compared to CBT (n = 35; SERFATY1999). I
There is limited evidence to suggest that inpatient treatment at a national tertiary eating disorder centre is less acceptable than tertiary specialist outpatient psychological treatments provided at the same centre (n = 90; CRISP1991). I
In children and adolescents, there is insufficient evidence to determine whether conjoint or separate family therapy are more, or less, acceptable (n = 64; EISLER2000, ROBIN1999). I
In children and adolescents, there is insufficient evidence to determine whether the addition of body awareness therapy to family therapy is more, or less, acceptable compared to family therapy alone (n = 33; WALLIN2000). I
6.2.4 Outpatient psychological treatments after weight restoration in hospital Inpatient treatments aim to return body weight to a healthy level but even when this occurs patients remain very vulnerable to subsequent weight loss. Outpatient psychological treatments are offered to prevent this.
6.2.4.1 Psychological treatments reviewed The following treatments were included: • Cognitive behaviour therapy (CBT)
96 CONFIDENTIAL Second draft for consultation September 2003
• Family therapy
• Dietary counselling
• Supportive therapy.
6.2.4.2 Studies considered The review team conducted a new systematic search for RCTs of outpatient psychological treatments after weight restoration in hospital in people with anorexia nervosa. Three small trials were included (GEIST2000, PIKESUBMIT, RUSSELL1987), providing data on 138 participants ranging in age from adolescents to adults.
Full details of studies included and excluded from the guideline are given in Appendix 17.
6.2.4.3 Evidence statements4 The statements in this section relate to adults, except where explicitly stated.
Effect of treatment on symptoms (weight and/ or proportion recovered or relapsed) There is limited evidence that individual supportive psychotherapy is superior in terms of weight gain when compared to family therapy in a subgroup of adults with anorexia nervosa (with adult onset) at one year post- treatment follow-up (n = 21; RUSSELL1987). I
There is limited evidence that for patients with an age of onset below 19 and with an illness duration of less than 3 years, family therapy focused explicitly on eating disorders is superior in terms of weight gain and the proportion classed as recovered when compared to individual supportive psychotherapy at one year post-treatment follow-up (n = 21; RUSSELL1987). I
There is limited evidence suggesting that CBT is superior to dietary counselling in terms of proportion recovered and relapse rates, after weight restoration in hospital (n = 33; PIKESUBMIT). I
In children and adolescents, there is insufficient evidence to determine the efficacy of family group education versus conjoint family therapy following partial weight restoration in hospital (n = 25; GEIST2000).
Acceptability of treatment
4 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
97 CONFIDENTIAL Second draft for consultation September 2003
There is limited evidence suggesting that CBT is more acceptable to people with anorexia nervosa when compared to dietary counselling, after weight restoration in hospital (n = 33; PIKESUBMIT). I
In children and adolescents, there is insufficient evidence to determine the acceptability of family group education versus conjoint family therapy following partial weight restoration in hospital (n = 25; GEIST2000).
6.2.5 Different types of psychological inpatient treatment regime There has been some interest in the question of whether different types of inpatient regimes are more efficient than others in terms of achieving short- term weight restoration.
6.2.5.1 Inpatient psychological treatments reviewed All inpatient treatments involving people with anorexia nervosa were considered in this section. These included specific forms of psychotherapy such as operant conditioning, and other behavioural regimes and psychodynamic psychotherapy, as well as (other) programmes focused on eating and weight gain.
6.2.5.2 Studies considered The review team conducted a new systematic search for different types of psychological inpatient treatment regime in people with anorexia nervosa. Because of the difficulties associated with this type of treatment and the paucity of RCT data, lower levels of evidence were examined. Thus, two RCTs (ECKERT1979, WEIZMAN1985) and two non-randomised controlled studies were included (HERZOG1996, SOLANTO1994,), providing data on 374 participants ranging in age from adolescents to adults.
Full details of studies included and excluded from the guideline are given in Appendix 17.
6.2.5.3 Evidence statements5
Effect of treatment on symptoms (weight gain) There is limited evidence (from one non-randomised study) to suggest that an inpatient programme with an explicit focus on changing eating disorder symptoms and weight is superior to a programme without this focus, in terms of producing short term weight gain (n = 34; HERZOG1996). IIa
There is very limited evidence from one small non-randomised study suggesting that increasing the 4-day criterion weight gain from 0.4 to 0.5 kg in
5 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
98 CONFIDENTIAL Second draft for consultation September 2003
a behavioural contract produces more rapid weight gain (n = 53; SOLANTO 1995). IIb
Acceptability of treatment There is insufficient evidence to determine whether any one form of inpatient treatment is more, or less, acceptable to people with anorexia nervosa when compared with another form of inpatient care. I-IIb
6.2.6 Psychological treatments as adjuncts to inpatient treatment Specialist inpatient treatment programmes for anorexia nervosa typically consist of multiple components, with the aim of providing comprehensive package of inpatient care. The relative importance and efficacy of these different components has so far received little research attention.
6.2.6.1 Psychological treatments reviewed The following treatments, as adjuncts to inpatient treatment, were included:
• Behaviour therapy (systematic desensitisation)
• Relaxation training
• Social skills training.
6.2.6.2 Studies considered The review team conducted a new systematic search for RCTs of psychological treatments as adjuncts to inpatient treatment in people with anorexia nervosa. Two small trials (GOLDFARB1987, PILLAY1981) were included, providing data on 41 participants ranging in age from adolescents to adults.
Full details of studies included and excluded from the guideline are given in Appendix 17.
6.2.6.3 Evidence statements6
Effect of treatment on symptoms and acceptability There is insufficient evidence to draw any conclusions from these studies, as to the efficacy of behaviour therapy (systematic desensitisation), relaxation training or social skills training as an adjunct to inpatient treatment (n = 41; GOLDFARB1987, PILLAY1981). I
6 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
99 CONFIDENTIAL Second draft for consultation September 2003
6.2.7 Special considerations in the management of children and adolescents There is a small but consistent evidence base which indicates that family based treatments are important in the treatment of adolescents with anorexia nervosa. Consensus as to what constitutes the best form of family intervention has not been achieved but two factors stand out as potentially important: • Family interventions should have a focus on the eating disorder and how this impacts on family relationships, emphasising in the early stages of treatment the necessity for parents to take a central role in supporting their child’s efforts to eat
• Both separated (parents and patient meet separately with the therapist) and conjoint forms (parents and patient together with therapist) of family therapy may be beneficial.
Establishing a collaborative working relationship with families with a young person with anorexia nervosa presents a particular challenge that requires time and expertise to balance the competing needs of different family members. However, whilst there is an emphasis on family interventions the young person’s individual rights and responsibilities should not be overlooked. Issues such as confidentiality and consent must be considered carefully and not simply overridden by clinicians or parents. For this reason, young people should be offered individual appointments with a therapist separate from those with their family members or carers. For children and adolescents it is also particularly important to ensure adequate physical monitoring and rapid commencement of treatment.
6.2.8 Clinical summary Various forms of psychological treatment are associated with improvements in terms both of weight gain and recovery by the end of treatment compared to ‘standard care’ for certain populations. However, the long-term benefits may not be sustained. For patients not requiring emergency admission to hospital, outpatient psychological treatment may be as, or more, effective than admission. For those admitted to hospital, no particular psychological treatment regime, either as a central component of the treatment programme or as an adjunct has been shown to have beneficial effects.
There is limited evidence that family interventions focused explicitly on eating disorders may be of specific benefit to younger people, but there is insufficient evidence to determine whether conjoint (i.e., patient and parents meet together) or separated forms of family therapy (i.e., therapist meets patient and parents separately) are more effective.
6.2.9 Clinical practice recommendations
Common elements of the psychological treatment of anorexia nervosa
100 CONFIDENTIAL Second draft for consultation September 2003
6.2.9.1 Therapies to be considered for the psychological treatment of anorexia nervosa include cognitive analytical therapy (CAT), cognitive behaviour therapy (CBT), interpersonal psychotherapy (IPT), focal psychodynamic therapy and family interventions focused explicitly on eating disorders. (C)
6.2.9.2 Patient and, where appropriate, carer preference, should be taken into account in deciding which psychological treatment is to be offered. (C)
6.2.9.3 The aims of psychological treatment should be to reduce risk, encourage weight gain and healthy eating, to reduce other eating- disorder related-symptoms and to facilitate psychological and physical recovery. (C)
Outpatient psychological treatments for adults
6.2.9.4 Most adults with anorexia nervosa should be managed on an outpatient basis with psychological treatment (with physical monitoring) provided by a healthcare professional competent in the psychological treatment and assessment of physical risk of people with eating disorders. (C)*
6.2.9.5 Outpatient psychological treatment and physical monitoring for anorexia nervosa should normally be of at least 6 months’ duration. (C)
6.2.9.6 For adults with anorexia nervosa, if during outpatient psychological treatment there is significant deterioration, or the completion of an adequate course of outpatient psychological treatment does not lead to any significant improvement, more intensive forms of treatment (e.g., a move from individual therapy to combined individual and family work; or day-care, or inpatient care) should be considered. (C)
6.2.9.7 Dietary counselling should not be provided as the sole treatment for anorexia nervosa. (C)
Psychological aspects of inpatient care
Psychological treatment is often a key element of an inpatient stay but evidence for what kind of treatment or approaches to treatment are effective is limited.
6.2.9.8 For inpatients with anorexia nervosa, a structured symptom-focused treatment regime with the expectation of weight gain should be
101 CONFIDENTIAL Second draft for consultation September 2003
provided in order to achieve weight restoration. It is important to carefully monitor the patient’s physical status during refeeding. (C)
6.2.9.9 Psychological treatment should be provided which has a focus both on eating behaviour and attitudes to weight and shape, and wider psychosocial issues with the expectation of weight gain. (C)
6.2.9.10 Rigid inpatient behaviour modification programmes should not be used in the management of anorexia nervosa. (C)
Post-hospitalisation psychological treatment for adults with anorexia nervosa
For patients with anorexia nervosa following discharge from hospital it is usually necessary to extend the duration of psychological treatment over that normally provided to those who have not been hospitalised.
6.2.9.11 Following inpatient weight restoration people with anorexia nervosa should be offered outpatient psychological treatment that focuses both on eating behaviour and attitudes to weight and shape, and wider psychosocial issues with regular monitoring of both physical and psychological risk. (C)
6.2.9.12 The length of outpatient psychological treatment and physical monitoring following inpatient weight restoration should typically be of at least 12 months’ duration. (C)
Children and adolescents with anorexia nervosa
Special considerations are needed in the treatment of children and adolescents, of particular importance is the involvement of families and other carers. Support from education and peers may also play a role in recovery and clinicians will often need to liase with schools over involvement in physical education and sitting examinations.
6.2.9.13 Family interventions focused explicitly on eating disorders should be offered to children and adolescents with anorexia nervosa. (B)*
6.2.9.14 Children and adolescents with anorexia nervosa should be offered individual appointments with a healthcare professional separate from those with their family members or carers. (C)
6.2.9.15 The therapeutic involvement of siblings and other family members should be considered in all cases because of the effects of anorexia nervosa on other family members. (C)
102 CONFIDENTIAL Second draft for consultation September 2003
6.2.9.16 In children and adolescents with anorexia nervosa the need for inpatient treatment and the need for urgent weight restoration should be balanced alongside the educational and social needs of the young person. (C)
103 CONFIDENTIAL Second draft for consultation September 2003
6.3 Pharmacological interventions
6.3.1 Introduction A diverse network of neurotransmitters and neurohormones are involved in the central and peripheral control of appetite and satiety. A variety of drugs which act on various receptors within these pathways have been examined in the treatment of anorexia nervosa.
6.3.2 Current practice Antidepressant drugs are often used to treat the depressive symptoms in anorexia nervosa and their effects on weight gain have also been studied. Medications are also used to treat comorbid conditions such as major depressive disorder and obsessive-compulsive disorder. Antipsychotic drugs or minor tranquillisers or antihistamines are frequently used symptomatically to reduce the high levels of anxiety present with anorexia nervosa, but are not recommended for the promotion of weight gain.
However, an evidence base for current practice is lacking. There are few studies on which to base clinical decisions and the studies are of low statistical power. There is also doubt about generalising from the patient samples studied. Drugs are not as acceptable or as well tolerated as psychotherapy in this patient group (Treasure, 1998). Only short-term effects have been studied and the outcome measures used are often not comparable to those used in psychotherapy studies. Compromised nutritional status may also affect the mechanism of drug action and this is rarely considered in studies. For example, there is some evidence that antidepressants have lower efficacy in the context of low levels of oestrogen (Halbreich & Kahn, 2000). They may be also less effective if tryptophan levels are altered.
Because of the complications of starvation, vomiting, dehydration and over hydration in this clinical group, there may be problems in terms of pharmacokinetics (i.e., drug absorption and toxicity). Extremely malnourished patients and those with electrolyte abnormalities are at risk of cardiac complications. Drugs with cardiac side effects should be used with caution.
6.3.3 Pharmacological treatment The aim of pharmacological treatment in people with anorexia nervosa is to produce weight gain and improve their quality of life or to alleviate some of the comorbidity such as depression, anxiety or obsessive compulsive features. Three main classes of drugs have been considered in the treatment of anorexia nervosa, these are antidepressants, antihistamines and antipsychotics.
6.3.3.1 Drugs reviewed
104 CONFIDENTIAL Second draft for consultation September 2003
The following drugs were included: • Antidepressants
• Tricyclic antidepressants (amitriptyline, clomipramine)
• SSRIs (fluoxetine, citalopram)
• Antihistamines (cyproheptadine)
• Antipsychotics (pimozide, sulipride)
Drugs that have had their licences withdrawn from the UK were not included in the guideline. Although there are studies assessing the use of lithium carbonate, these were not included, as it was felt inadvisable to use this drug given the potential risk of toxicity in these patients.
6.3.3.2 Studies considered A new search was made for RCTs examining drugs used in the treatment of anorexia nervosa. One trial of both an antidepressant and an antihistamine (HALMI1986), five trials of antidepressants (ATTIA1998, BIEDERMAN1985, FASSINO2002, KAYE2001, LACEY1980), one trial of an antihistamine (GOLDBERG1980), and three trials of antipsychotics (RUGGIERO2001, VANDEREYCKEN1982, VANDEREYCKEN1984) met the eligibility criteria set by the GDG. Thus, ten RCTs involving 413 adult participants were included in this section.
Of the eleven trials, four trials involved a comparison of a tricyclic antidepressant (amitriptyline, clomipramine) with placebo (BIEDERMAN1985, HALMI1986, KAYE2001, LACEY1980), one compared a SSRI antidepressant (fluoxetine) with placebo (ATTIA1998), and one compared a SSRI antidepressant (citalopram) with a wait-list control (FASSINO2002). Two trials involved a comparison of an antihistamine with placebo (GOLDBERG1980, HALMI1986). Two trials compared an antipsychotic with placebo (VANDEREYCKEN1982, VANDEREYCKEN1984), and one compared an antipsychotic with both another antipsychotic and with an antidepressant (RUGGIERO2001).
Full details of studies included in the guideline and reasons for excluding studies are given in Appendix 17.
6.3.3.3 Evidence statements7
7 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
105 CONFIDENTIAL Second draft for consultation September 2003
The data were analysed by combining all studies, irrespective of the class of antidepressant used. However, where the data are from a single drug, the class and name of drug are reported.
Effect of treatment on weight gain There is evidence suggesting that it is unlikely there is a clinically significant difference between antidepressant drugs and placebo on weight gain by the end of multi-modal inpatient treatment (n = 146; ATTIA1998; BIEDERMAN1985; HALMI1986; LACEY1980). I
There is insufficient evidence to determine whether antipsychotics or antihistamines have any impact on weight compared with placebo during multi-modal inpatient treatment. I
There is insufficient evidence to determine whether there is any difference between antipsychotics and antidepressants with regard to weight gain. I
Effect of treatment on relapse/ clinical deterioration There is limited evidence suggesting that there is a clinically significant difference between an SSRI (fluoxetine) and placebo with fewer patients deteriorating clinically (which for the majority of patients was defined as a worsening or no improvement in symptoms) following inpatient weight restoration if given fluoxetine for one year (N = 1; n = 35; RR = 0.45; 95% CI, 0.23 to 0.86). I
Acceptability of treatment There is insufficient evidence to determine whether antidepressants, antipsychotics, or antihistamines are more, or less, acceptable to people with anorexia nervosa when compared to placebo or wait-list control. I
Tolerability of treatment There is insufficient evidence to determine whether antidepressants, antipsychotics, or antihistamines produce a great risk of side effects in people with anorexia nervosa when compared to placebo. I
6.3.4 Special considerations in the management of children and adolescents It is uncertain whether any of the above findings can be generalised to children and adolescents. In general there are safety data available for sertraline and amisulpride used for other conditions in the under 18 group, but these drugs have not been studied in anorexia nervosa. There is no evidence specifically addressing the use of drugs in the child and adolescent age group.
6.3.5 Clinical summary There is no evidence that drug treatment (antidepressants [tricyclics and SSRIs]), conventional antipsychotics, antihistamines) has additional benefit on
106 CONFIDENTIAL Second draft for consultation September 2003
weight gain in people undergoing multi-facetted inpatient treatment. There is limited preliminary evidence from one small trial that fluoxetine may reduce the likelihood of deterioration after inpatient weight restoration. It remains to be established whether or not new compounds (e.g., atypical antipsychotics) or new settings (e.g., day hospitals as a supplement to outpatient therapy) or new specifications (matching drug to clinical phenotype or endophenotype) will be of benefit. A further complication in interpreting these studies is that a number of the secondary features of anorexia nervosa which may respond to medication may also improve as the patient gains weight (e.g., depressed mood).
6.3.6 Clinical practice recommendations
6.3.6.1 Medication should not be used as the sole or primary treatment for anorexia nervosa. (C)
6.3.6.2 Caution should be exercised in the use of medication for comorbid conditions such as depression or obsessive-compulsive disorder, as many of these features may resolve with weight gain alone. (C)
6.3.6.3 When medication is used to treat people with anorexia nervosa, the side effects of drug treatment, (in particular, cardiac side effects), should be carefully considered because of the compromised cardiovascular function of many people with anorexia nervosa. (C)
6.3.6.4 Healthcare professionals should be aware of the risk of drugs which prolong the QTc interval: for example, antipsychotics, tricyclic antidepressants, macrolide antibiotics, and some antihistamines. (C)
6.3.6.5 In patients with anorexia nervosa with borderline/prolonged QTc interval the prescription of these drugs with side effects that may compromise cardiac functioning should be avoided. (C)
6.3.6.6 If the prescription of medication that may compromise cardiac functioning, is essential ECG monitoring should be undertaken. (C)
6.3.6.7 All patients with a diagnosis of anorexia nervosa should have an alert placed in their prescribing record concerning the risk of side effects. (C)
6.4 Management of physical aspects
6.4.1 Introduction Identification of those patients with short-term risk of serious harm or death is clearly important in clinical practice. Recommendations as to how to assess, what to monitor and when to intervene are therefore vital.
107 CONFIDENTIAL Second draft for consultation September 2003
The vast majority of longer-term follow-up studies indicate that people with anorexia nervosa have an almost 10-fold risk of dying compared to healthy people the same age and sex. Standardized mortality rates range between 4.71 and 12.82 (Nielsen et al., 1998). Mortality in eating disorders is predominately related to malnutrition, methods of weight control and suicide. Among studies in which cause of death is documented, 54% died of eating disorder complications, 27% committed suicide and the remaining 19% died of unknown or other causes (Nielsen, 2001).
6.4.2 Current practice There is limited evidence as to how and where risk should be managed. Opinion and practice varies between (and within) countries, centres and clinicians. This is the case both in terms of threshold for hospital admission and the goals of the admission. For example, the APA guidelines (American Psychiatric Association, 2000) recommend admission to hospital when a BMI is less than 16 kg/m2 or weight loss greater than 20%. In both the US and Australia, patients are generally admitted for short-term medical stabilisation on a medical or paediatric ward. In the United Kingdom, patients at low weight are frequently managed in an outpatient setting in specialist services. In these instances there is a higher threshold for inpatient treatment with admission often not occurring until the patient’s BMI falls below 13 kg/m2. However managing these low weight patients in an outpatient setting can be hazardous and should rarely be done without specialist advice. In the majority of inpatient specialist services in the UK, the goal is full weight recovery and so admissions are longer term. Admission to either a medical/ paediatric or general psychiatric unit may occur but this varies with availability of services.
6.4.3 Physical risk reduction and monitoring Applying information about the best evidence to a specific patient’s problem is not easy as the clinical picture is diverse physically, psychologically and socially. The patient’s problems have to be clearly defined and placed within the context of a clinical risk assessment. This needs to include an assessment both of the acute risk and the longer-term prognosis.
Decisions on short-term risk involve a combined assessment of the physical risk and the person’s psychological capacity to consent to treatment, taking into account the possible resources of motivation and psychosocial support. A diagram is provided (see Appendix 7) as a simple guide to medical practitioners and other members of the multidisciplinary team as a decision aid when evaluating this acute risk. Body mass index is a better marker than weight alone as a proxy measure of physical risk but a rigid cut off point is less good for the extremes of height as the relationship is non-linear. Children have smaller fat stores than mature women and so medical complications occur with less weight loss. Bulimic features or refusal to drink also increase
108 CONFIDENTIAL Second draft for consultation September 2003
the risk. In turn, these medical markers interact with a variety of clinical and psychosocial factors. High physical risk is often associated with an impairment of capacity for the consent to treatment.
Because of the paucity of data and the nature of the issue under review, the GDG chose to use an informal consensus process (see chapter 3, section 3.4.7. for details) to address questions related to risk reduction and monitoring. The review team conducted a systematic search for all available evidence relating to the issue. A number of studies met the GDG’s eligibility criteria, including two systematic reviews of anorexia nervosa and mortality, and outcome at follow-up (Nielsen, 2001; Steinhassuen, 2002). Long-term follow-up studies on patients suffering from anorexia nervosa were also considered, and where possible predictors and causes of death were noted. This is an area in which there is limited research. Exact causes of death are rarely described. The majority of studies documented cause of death as being as a result of starvation in anorexia nervosa. In those studies where death certificates were examined, again the exact precipitant of death was not always established. There are few post mortem results documented.
Physical factors associated with higher mortality include severity of weight loss, over-activity and vomiting, bulimia and purging (Nielson, 2000; Steinhausen, 2002). A BMI less than 13 kg/m2 in adults is of prognostic significance as it indicates a greater risk for mortality (Casper, 1996, Heberand et al., 1997). In the latter study, only 7 out of the 14 patients with a BMI of 11 kg/m2 at referral survived (Hebebrand, 1997). Data on malnourished females in famine indicate that with a BMI less than 11 kg/m2, risk of mortality increases sharply (Collins, 1995). Prolonged QT intervals may predispose a person to life-threatening arrhythmias, and might be responsible for cases of sudden death. Furthermore, prolonged QT intervals in ECG were recorded 7 days before sudden death (Isner, 1985). Abnormally low serum albumin levels and low weight are the best variables to predict a lethal course (Herzog et al., 1997).
In the studies documenting mortality in patients with anorexia nervosa, causes of death vary. Causes include dehydration, electrolyte (particularly hypokalaemia) and metabolic complications, infections (bronchopneumonia and sepsis) and cardiac complications (see Neumarker, 1997, for review). Comorbid alcoholism has also been shown to be associated with increased mortality (Keel et al., 2003). Rupture / perforation of the gastrointestinal tract has been less frequently described (Zipfel, 2000)
Few studies have reported exclusively on the mortality in adolescents. There is a suggestion that the mortality rate is lower. This may reflect the fact that chronicity and mortality increase with increasing age (Steinhausen, 2002). BMI measures in children and younger adolescents are an inadequate reflection of physical reserve. Weight centiles or percentage weight for height
109 CONFIDENTIAL Second draft for consultation September 2003
give a more accurate guide to the degree of weight loss. Rapid weight loss in children is potentially more dangerous than in adults, and children are known to dehydrate more quickly (Irwin, 1984). Clinical judgement and physical examination may be a better indicator than serum electrolyte levels as to the need for rehydration (Nicholls & Stanhope, 2000).
6.4.4 Clinical summary
Low weight (BMI less than 13 kg/m2), dehydration and electrolyte abnormalities indicate an increased risk of mortality. Cardiac arrhythmias and low serum albumin and glucose are of particular concern.
6.4.5 Clinical practice recommendations
Managing weight gain
6.4.5.1 In most patients with anorexia nervosa an average weekly weight gain of 0.5-1 kg in inpatient settings and 0.5 kg in outpatient settings should be an aim of treatment. This requires about 3,500 to 7000 extra calories a week. (C)
6.4.5.2 Healthcare professionals should ensure for children and adolescents with anorexia nervosa that once a healthy weight is reached the increased energy and necessary nutrients are available in the diet to support growth and development. (C)
6.4.5.3 Regular physical monitoring and in some cases treatment with multi- vitamin/ multi-mineral supplement in oral form is recommended for people with anorexia nervosa during both inpatient and outpatient weight restoration. (C)
6.4.5.4 In the nutritional management of children and adolescents with anorexia nervosa, carers need to be included in any dietary education or meal planning. (C)
6.4.6 Long term risk and its management Apart from the concern of immediate physical risk in patients with anorexia nervosa, the long-term physical consequences are considerable. For the purposes of this section, the GDG chose to focus on the effects on the skeletal system and on reproduction. Gastrointestinal and dental complications are largely as a result of vomiting and laxative abuse. These are described in the chapter on bulimia nervosa. It should be noted however that many of these complications are exacerbated in patients with low weight and muscle weakness. Other consequences have also been described (e.g., cardiac, dermatological, haematological), but these will not be covered in this guideline.
110 CONFIDENTIAL Second draft for consultation September 2003
The development of osteopenia and osteoporosis is a serious and long-term consequence of starvation. Oestrogen deficiency, malnutrition, low body mass and hyperactivity all play a part in this development. This section will consider evidence for the effect of weight gain and medication in the management of this complication. In children and adolescents, weight loss has particularly serious implications. Sixty percent of bone accretion occurs during puberty (Golden, 1992). Failure of this accretion thus compounds bone loss in children and adolescents with anorexia nervosa. Growth retardation is a further complication in children and adolescents (Russell, 1985, Danziger, 1994) and failure to grow may indicate nutritional deficiencies. It is thought that with weight gain “catch-up growth” can occur up until fusion of the epiphyses (Nicholls & Stanhope, 2000). Regular monitoring of height as well as weight is important in children and adolescents with anorexia nervosa. Recommended weight ranges have to be regularly adjusted to take into account changes in height and age. Hormonal treatments, although used, have not been evaluated in adolescents with anorexia nervosa. There is a risk that oestrogen administration may cause premature epiphyseal fusion and growth stunting (Nicholls & Stanhope, 2000).
With regard to the reproductive system, there are several areas of concern: infertility, persistent amenorrhoea and oligomenorrhoea, and polycystic ovaries. In adolescent anorexia nervosa there is a risk of pubertal delay and ultimately arrested pubertal development (Russell, 1985). The effect on fertility and pregnancy will be described below.
Dental erosion is the most common oral problem in patients with eating disorders that engage in self-induced vomiting. A discussion of the issues concerning dental complications can be found in the chapter on bulimia nervosa (section 7.5.2).
Osteoporosis People with anorexia nervosa have reduced bone mineral density (BMD) (Bachrach, Guido, Katzman, Litt & Marcus, 1990) which is associated with an increased fracture rate (Rigotti, Neer, Skates, Herzog & Nussbaum, 1991; Vestergaard et al., 2002; Lucas, Melton, Crowson & O'Fallon, 1999; Soyka, Grinspoon, Levitsky, Herzog & Klibanski, 1999) and long term disability such as pain, kyphosis and loss of height. Osteoporosis is manifest in some people after a year of illness and the severity gradually increases over time if the illness remains untreated. Fractures may occur after a few years of illness.
The factors that predict bone density in the majority of studies include: duration of amenorrhoea (Biller et al., 1989; Seeman, Szmukler, Formica, Tsalamandris & Mestrovic, 1992;), BMI, or some other variable which reflects body composition (Grinspoon et al., 2000; Some studies find that age at
111 CONFIDENTIAL Second draft for consultation September 2003
menarche is also a predictor (Grinspoon et al., 2000). Oestrogen use does not predict density in larger studies (Grinspoon et al., 2000).
There have been several longitudinal studies that have examined changes in bone density over time. In the majority of cases the people are also engaged with treatment, commonly nutritional and psychological interventions. In some studies additional treatments such as fluoride and hormone replacement have been added. It is difficult to compare between studies as the presentation and type of data differs between studies, as does the case mix.
Adolescents. One of the largest studies in adolescents is that of Castro and colleagues (Castro et al., 2001). They have an early intervention service with a young (14.4 years) and moderately severe (BMI 15 kg/m2) group. They found that change over time depends on the severity of the osteoporosis at baseline and the degree of recovery made over the period of follow-up. The group with established osteopenia at baseline (i.e., z < -1.0) had an increase in bone density of over 9% at spine (5.7% in the femur) over the follow-up interval (1.3 years). Adults. In adults (mean 22 years) with a severe illness (mean BMI 13.9 kg/m2) followed over 2 years, when BMI increased to 17.1 kg/m2 there was an overall 2% increase in bone density which related to weight gain. The subgroup that had full recovery (weight gain and menstruation) had the largest increase in bone density (Iketani, et al., 1995).
An additional study (Zipfel et al., 2001), in adults (21 years) with BMI 14 kg/m2 found similar results.
Current Clinical Practice. The most effective treatment/ preventative agent for osteoporosis in anorexia nervosa is not yet known. Adequate nutrition and weight are the most relevant factor but in some cases this is difficult to implement in the long term. Therefore, there has been interest in replacing some factors of relevance to bone turnover. These include oestrogen (in the form of hormone replacement therapy or the contraceptive pill) which inhibits bone resorption and Vitamin D, calcium, and insulin-like growth factors (Grinspoon, 1997; Grinspoon, 2002; DHEA; Gordon, 2002. Bisphosphonates and fluoride have also been tried (Maugars et al., 1996).
6.4.6.1 Treatments reviewed The following treatments were included: • De hydroepiandrosterone (DHEA)
• Insulin-like growth factor (IGF-I)
• Oestrogen supplementation.
112 CONFIDENTIAL Second draft for consultation September 2003
6.4.6.2 Studies considered The review team conducted a new systematic search for studies examining factors associated with the management of osteoporosis in people with anorexia nervosa. Five RCTs met the eligibility criteria set by the GDG (GORDON1999, GORDON2002, GRINSPOON1996, GRINSPOON2002, KLIBANSKI1995), involving 207 participants.
Of the five studies included, one compared oestrogen supplementation with assessment only (KLIBANSKI1995), one compared different doses of DHEA (GORDON1999), one compared DHEA with hormone replacement therapy (GORDON2002), one compared two different doses of IGF-I with placebo (GRINSPOON1996), and one compared IGF-I, oestrogen supplementation, placebo and the combination of IGF-I and oestrogen (GRINSPOON2002).
6.4.6.3 Evidence statements8 There is insufficient evidence to determine whether oestrogen supplementation improves bone density by the end of treatment. I
There is insufficient evidence to determine whether oral DHEA improves bone density compared to hormone replacement therapy by the end of treatment. I
There is limited evidence that both IGF-I alone and the combination of IGF-I and an oral contraceptive may improve bone density:
• There is limited evidence suggesting that there is a clinically significant difference between rhIGF-I and placebo with IGF being superior in terms of bone turnover/ density by the end of treatment (GRINSPOON1996, GRINSPOON2002). I
• There is limited evidence suggesting that there is a clinically significant difference between rhIGF-I + oral contraceptive and placebo with the combination being superior in terms of bone density by the end of treatment (GRINSPOON2002). I
There is some evidence that the combination of rhIGF-I and an oral contraceptive is produces a better outcome than rhIGF-I alone.
There is insufficient evidence to determine whether oestrogen supplementation is more or less acceptable to people with anorexia nervosa when compared to no oestrogen supplementation. I
8 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
113 CONFIDENTIAL Second draft for consultation September 2003
There is insufficient evidence to determine whether oral DHEA is more or less acceptable to people with anorexia nervosa when compared to hormone replacement therapy. I
Full recovery from anorexia nervosa with weight gain and return of menstruation leads to a marked increase in bone density (the rate depends on the initial level, but can be as much as 5% or more a year). However in those that remain under weight with amenorrhoea bone loss continues. III
In adolescents with a good outcome and low initial bone density the increase in BMD is 4 fold that of normal adolescents. III
6.4.7 Clinical summary Bone loss is a serious problem in anorexia nervosa with serious long-term consequences. Weight restoration is associated in adolescents with important gains in bone density. Oral oestrogen and oral DHEA do not appear to have a positive impact on bone density and hormone replacement therapy is not recommended in children and adolescents as it may cause premature fusion of the bones, and high impact exercise is associated with an increased risk of fracture in anorexia nervosa. rhIGF-I, alone or in combination with an oral contraceptive, is associated with improvements in bone metabolism and bone but intensive clinical monitoring is necessary and this treatment should only be given in specialist centres. The long-term effect is uncertain.
6.4.8 Clinical practice recommendations
6.4.8.1 Oestrogen administration should not be used to treatment bone density problems in children and adolescents as this may lead to premature fusion of the epiphyses. (C)
6.4.9 Other physical interventions Malnutrition is a core feature of anorexia nervosa. Refeeding is a necessary component of treatment but is not sufficient. The approach to refeeding varies between centres and countries. There is debate about the setting, means and rate of weight gain and limited evidence to support different views. In some centres the calorie deficit is made up with food, given as normal, albeit larger, meals or snacks. In other centres liquid foods can be used to supplement or replace some or all of the meals. An alternative, which is not used frequently within the UK, is nasogastric feeding. Rarely percutaneous endoscopic gastrostomy (PEG) or total parenteral nutrition (TPN) has been used. These interventions are only used when patients are not able to cooperate with oral refeeding and there is concern about physical risk. In these circumstances legal and ethical considerations need to be addressed.
A number of complications can occur during refeeding. This is a high risk period for biochemical abnormalities. People at most risk are those with a
114 CONFIDENTIAL Second draft for consultation September 2003
BMI less than 12 kg/m2, those who vomit, abuse laxatives and binge, and those with concurrent physical conditions. Physical monitoring is necessary during periods of refeeding. A range of electrolyte disturbances can occur during re-feeding, which are sometimes referred to collectively as the “refeeding syndrome”. Hypophosphataemia may develop rapidly during refeeding; if severe, it can cause cardiac and respiratory failure, delirium and fits. Ingestion of large quantities of carbohydrates, during rapid re-feeding, may result in a precipitate drop in serum phosphate levels.
Certain vitamin or mineral deficits have come under close scrutiny. For example, the similarity between some of the symptoms of zinc deficiency and the symptoms of anorexia nervosa has led to an interest in zinc replacement. A proportion of patients with anorexia nervosa may be deficient in thiamine, riboflavin, Vitamin C and Vitamin D. The clinical significance of many of these deficiencies is unclear but it is usual for there to be a general rather than a specific deficit and therefore a multi-vitamin/multi-mineral supplement in oral form (e.g., suitable preparations include Sanatogen Gold (non-NHS), Forceval 1-2 or Seravit capsules daily. Care should be taken to prevent the risk of vitamin A and D toxicity from excessive use of supplements (Department of Health, 1991).
There has been some research into the nutritional management of anorexia nervosa but all of the studies on nutritional approaches are quasi- experimental and small, and the studies on zinc are small.
There is controversy, but little evidence, about the appropriate time course used to achieve the goals of treatment. Some argue that it is important to restore normal weight as soon as possible, others argue for a slower increase in weight. The standard rate of weight gain in the UK is 0.5 – 1 kg per week. The majority of people with anorexia nervosa are treated as outpatients but for the small number who require admission most specialist inpatient units aim to discharge people once weight is fully recovered. However, there is an alternative view that full weight recovery can be achieved by outpatient or day patient interventions following brief inpatient refeeding, and that discharge at an intermediate weight may provide an alternative.
There is also the argument that weight gain is only one outcome of interest. Another goal is to ensure that eating behaviour is normalised and maintained after discharge, that abnormal weight and shape cognitions are normalised and that there is improved quality of life in people with anorexia nervosa.
Physical therapies have been used in some European countries (e.g., Belgium and Norway), which have a long tradition of integrating physical therapies into psychotherapy. Physical rehabilitation programmes including riding, climbing etc have been used. This has not been the tradition in the UK. Indeed
115 CONFIDENTIAL Second draft for consultation September 2003
in many programmes in the UK exercise is restricted and bed rest is prescribed.
Feeding in the context of active resistance This section relates to those occasions where the individual requires restraining to allow the refeeding to take place. Feeding in the context of active resistance raises ethical, legal and clinical issues for all involved. The reporting of it in the general media is often inaccurate and emphasises the emotive “force feeding” and the suggested abuse of the individual's rights who is fed in this way. In the UK in the treatment of people with anorexia nervosa it is a very rare event and should only be done in a specialised treatment setting with access to the skills and experience required to manage it safely and effectively. It raises complex legal issues. It is interesting to note that all requests to the courts for permission to carry out this intervention in anorexia nervosa have been granted. It is undertaken using the Mental Health Act 1983, the Children Act 1989 and parental authority. Only a small proportion of patients who are admitted and treated compulsorily require this intervention. No studies have reported on the characteristics of those who are fed in this way, or have followed them up.
Methods of delivering this intervention vary across the country, as do the circumstances under which it is used. The risks associated with naso-gastric (NG) tube feeding, PEG, or spoon feeding, will be increased in the context of active physical resistance. Actions such as the pulling out the (NG) tube, interfering with or pulling out the PEG, and the physical condition of the patient increase the risk involved.
Clinical decisions as to when to start the intervention, how long to continue for, how often to repeat the intervention and when to stop are complex and difficult. There are further complications with children and adolescents where the urgency to intervene is increased because the low fat mass means that any weight loss is predominantly of lean tissue with higher medical consequences. As a result, children can become dangerously physically compromised. In addition, children and younger adolescents have a much greater tendency to restrict both fluid and food intake. The combination of these factors leads to a rapid breakdown of muscle tissue and dehydration, especially in pre-pubertal children.
6.4.9.1 Physical interventions reviewed The following interventions were included: • Nasogastric feeding
• TPN
• Zinc supplementation
116 CONFIDENTIAL Second draft for consultation September 2003
• Massage
• Exercise.
6.4.9.2 Studies considered The review team conducted a new systematic search for RCTs of physical interventions used in the treatment of anorexia nervosa. As there were few RCTs, the GDG elected to utilise lower levels of evidence where necessary. This included the APA guidelines and a recent expert review on the nutritional management of anorexia nervosa (EDSIG). Eight studies met the eligibility criteria set by the GDG (ARII1996, BIRMINGHAM1994, HART2001, KATZ1987, LASK1993, PERTSCHUK1981, ROBB2002, THIEN2000), involving 267 participants.
Of the eight studies, two involved nasogastric feeding (ARII1996, ROBB2002), one TPN (PERTSCHUK1981), three zinc (BIRMINGHAM1994, KATZ1987, LASK1993), one exercise (THIEN2000), and one massage (HART2001).
Full details of studies included in the guideline and reasons for excluding studies are given in Appendix 17.
6.4.9.3 Evidence statements9
Effect of treatment on weight gain There is limited evidence suggesting that there is a clinically significant difference between nasogastric feeding and 'standard care' with nasogastric feeding being superior in terms of weight gain by the end of multi-modal inpatient treatment (n = 116; ARII1996; ROBB2002). II
There is insufficient evidence to determine whether zinc supplementation has any impact on weight compared with placebo during multi-modal inpatient treatment. I
There is insufficient evidence to determine whether TPN has any impact on weight compared with placebo during multi-modal inpatient treatment. II
There is insufficient evidence to determine whether massage or exercise given in addition to ‘standard care’ have any impact on weight compared with ‘standard care’ alone by the end of treatment. I
Effect of treatment on symptoms of anorexia nervosa
9 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
117 CONFIDENTIAL Second draft for consultation September 2003
There is limited evidence suggesting that there is a clinically significant difference between massage and 'standard care' with massage being superior on EDI scores by the end of treatment (N = 1; n = 16; SMD = 1.06; 95% CI, 0.02 to 2.09). I
Acceptability of treatment There is insufficient or no evidence to determine whether nasogastric feeding or TPN are more, or less, acceptable to people with anorexia nervosa when compared to placebo or ‘standard care’. II
There is insufficient evidence to determine whether zinc supplementation is more, or less, acceptable to people with anorexia nervosa when compared to placebo or ‘standard care’. I
There is insufficient evidence to determine whether massage or exercise are more, or less, acceptable to people with anorexia nervosa when compared with ‘standard care’. I
Tolerability of treatment There is evidence from a retrospective chart review that TPN may produce a greater risk of side-effects than ‘standard care’:
• There is limited evidence suggesting that there is a clinically significant difference between TPN and ‘standard care’ with placebo being superior in terms of the number of people experiencing adverse events by the end of treatment (n = 22; PERTSCHUK1981). II
6.4.10 Clinical summary Nasogastric feeding can confer some benefit in terms of increased rate of weight gain or actual weight gain, as part of a treatment programme. There was insufficient evidence that either TPN or zinc supplementation confer any benefit in terms of weight gain. TPN appears to be associated with more adverse events in one small study. Some limited benefit, on symptoms but not on weight gain, has also been identified from one small trial investigating massage.
6.4.11 Clinical practice recommendations
6.4.11.1 Total parenteral nutrition should not be used for people with anorexia nervosa, unless there is a specific indication such as gastrointestinal dysfunction. (C)
6.4.12 Concurrent physical conditions
Diabetes
118 CONFIDENTIAL Second draft for consultation September 2003
There does not appear to be an increased occurrence of anorexia nervosa in Type 1 Diabetes, however the statistical power of the studies may be insufficient to rule this out (Nielsen, 2002). The mortality rate in 10 years of follow up of population based samples was found to be 2.2 (per 1000 person years) for Type 1 Diabetes, 7.3 for anorexia nervosa and 34.6 for concurrent cases (the standardized mortality rates were 4.1 for Type 1 Diabetes, 8.9 for anorexia nervosa, and 14.5 in concurrent cases) (Nielsen, 2002).
Close liaison and a shared knowledge base between the eating disorder and diabetes teams is essential in the management of anorexia nervosa with diabetes and they should have intensive regular physical monitoring as they are at high risk of complications and death.
Pregnancy It is unusual for people with anorexia nervosa to become pregnant although a small proportion have fertility treatment to conceive or relapse into anorexia nervosa during the pregnancy. In a follow up study of people with anorexia nervosa (n = 140) fertility was reduced to one third of the expected, the rate of prematurity was twice that expected and perinatal mortality was increased 6 fold (Brinch et al., 1988). In a follow up series of 66 women there was an increased miscarriage rate and more use of Caesarean sections and the offspring were more likely to be born prematurely and smaller (Bulik et al., 1999). On the other hand Steiner found no difference in weight gain and fetal weight in their sample (Steiner, Smith, Rosenkranz & Litt, 1991). In a study in which pregnant women with anorexia nervosa were followed prospectively the infants grew slowly in utero especially in the last trimester (Treasure & Russell, 1988).
Women with anorexia nervosa are at greater risk for premature offspring and those that are small for gestational age. There are case series that suggest that some women with anorexia nervosa have difficulty feeding their children (Russell, Treasure & Eisler, 1998; Wezel-Meijler & Wit, 1989) and that the child’s growth can be abnormal (Hodes, et al., 1997).
Management of laxative abuse The management of laxative abuse is covered in the bulimia nervosa chapter (section 7.5.2), as the management is essentially the same for both disorders.
6.4.13 Clinical practice recommendations
Managing risk
6.4.13.1 Healthcare professionals should monitor physical risk in patients with anorexia nervosa and where this leads to the identification of increased physical risk, the frequency and the monitoring and nature of the investigations should be adjusted accordingly. (C)
119 CONFIDENTIAL Second draft for consultation September 2003
6.4.13.2 People with anorexia nervosa and their carers should be informed if the risk to their physical health is high. (C)
6.4.13.3 The involvement of a physician or paediatrician with expertise in the treatment of physically at-risk patients with anorexia nervosa should be considered for all individuals who are physically at-risk. (C)
6.4.13.4 Pregnant women with either current or remitted anorexia nervosa may need more intensive prenatal care to ensure adequate prenatal nutrition and fetal development. (C)
Feeding against the will of the patient
6.4.13.5 Feeding against the will of the patient should be an intervention of last resort in the care and management of anorexia nervosa. (C)
6.4.13.6 Feeding against the will of a patient is a highly specialised procedure requiring expertise in the care and management of those with severe eating disorders and the physical complications associated with it. This should only be done in the context of the Mental Health Act 1983 or Children Act 1989. (C)
6.4.13.7 When making the decision to feed against the will of the patient the legal basis for any such action must be clear. (C)
6.5 Service interventions
6.5.1 Introduction The majority of people with eating disorders present first in primary care. However, it is probable that most patients presenting with anorexia nervosa are referred on to secondary care usually within mental health services, including both general mental health services and specialist eating disorder services. Specialist services for anorexia nervosa are distributed patchily (Royal College of Psychiatrists, 2001). A substantial proportion of tertiary specialist provision, in particular inpatient care, is in the private sector and many NHS patients are treated in private hospitals.
6.5.2 Current practice Most patients with anorexia nervosa receive treatment solely on an outpatient basis. However, a substantial minority receive inpatient treatment. The style and content of inpatient programmes for anorexia nervosa varies widely. Hospital admission may be aimed at ameliorating the effects of the illness on the patient’s physical state or at achieving progress toward full recovery. The
120 CONFIDENTIAL Second draft for consultation September 2003 former sometimes involves admission to general medical settings. The latter is usually attempted within psychiatric hospitals. It is generally held that it is preferable for admission to be to settings where the staff are experienced in the treatment of the disorder. Furthermore, the setting needs to be appropriate to the age of the patient. Special day programmes are being developed as an alternative mode of intensive treatment for anorexia nervosa.
6.5.3 Effective service configurations Although there is considerable variation in the pattern of service delivery there is a limited evidence base on which to develop effective patterns of service delivery. However, the principle that the right treatments should be offered to the right patients in the right setting if services are to be effective and cost effective, should guide service development.
This review seeks to examine whether any particular level of service provision is associated with better outcomes in anorexia nervosa. In addressing this issue it may be useful to differentiate between two types of hospital admission.
First, inpatient treatment of anorexia nervosa may be aimed at the stabilisation and rectification of the patient’s physical state. It is widely believed that such treatment can at times be life saving. However, such management has not been the subject of systematic comparative research and is unlikely to be.
Second, inpatient treatment aimed at helping the patient to progress toward full recovery is widely practiced and is more amenable to study.
The literature contains many descriptions of special inpatient treatment regimes for anorexia nervosa, and some outcome data are available (e.g., Bowers & Anderson, 1994). However, comparison between centres is difficult because of the likelihood that the case mix varies widely across studies. There is less data available about outpatient or day patient approaches. (Particular outpatient therapies are described elsewhere.) There is very little on the comparison between outpatient and day- or inpatient treatment or on the integration of different forms of care.
The St. Georges study (Crisp, Norton, Gowers, Halek, Yeldham, Levett & Bhat, 1991) is alone in attempting a comparison within a randomised controlled trial of hospital admission and two forms of outpatient treatment. (There was also an assessment only condition but many participants went on to receive treatment elsewhere thereby confounding the comparison.)
The study lacked power and had other difficulties. However, it did demonstrate that many patients made progress with fairly modest outpatient treatment over one year. (Gowers, Norton, Halek & Crisp 1994) Such
121 CONFIDENTIAL Second draft for consultation September 2003 outpatient treatment is widely practiced but poorly documented. In one study of a specialised secondary service nearly three quarters of adults with anorexia nervosa were managed without hospital admission (Palmer, Gatward, Black & Park, 2000).
Specialised day patient treatment for anorexia nervosa has been described in this country and abroad and positive outcomes are documented at least in the short term (Birchall, Palmer, Waine, Gadsby & Gatward, 2002; Gerlinghof, Backmund & Franzen, 1998; Olmsted, McFarlane, Molleken & Kaplan, 2001; Robinson, 2003; Zipfel et al., 2002). However, it is uncertain whether the patients described might otherwise have been admitted as inpatients or managed as outpatients. There are no randomised comparisons. One study suggests that the addition of a day programme may reduce the use of inpatient beds (Birchall, Palmer, Waine, Gadsby & Gatward, 2002). It is unlikely that day care could ever abolish the need for inpatient treatment altogether. Indeed, the relative effectiveness and cost effectiveness of the two forms of more intensive treatment have yet to be adequately studied.
It is widely believed that there may be benefits in the treatment of severe anorexia nervosa which has not responded to less intensive treatments within a specialised tertiary eating disorders service compared with less specialised secondary services. Both competence and confidence tend to develop in settings where such treatment is a regular and ongoing activity. However, there is a lack of studies that might provide evidence to support these views. One study of mortality in patients treated in two contrasting services suggested that there might be advantages associated with greater specialisation (Crisp, Callender, Halek & Hsu, 1992).
A single systematic review of inpatient versus outpatient treatment of anorexia nervosa was inconclusive because of lack of evidence (Meads, Gold & Burls, 2001).
6.5.4 Clinical summary In a minority of patients, admission to hospital may at times be necessary to stabilize the physical state or even save the life of severely physically impaired patients. Inpatient treatment aimed at recovery usually leads to weight gain at least where admission has been to a unit where such treatment is a regular activity. Such treatment may have lasting effects although weight loss is common after discharge. There is no unequivocal evidence that inpatient treatment confers long-term advantage except as a short-term life saving intervention in patients at high risk. However, inpatient treatment may well be a rational option for patients who have failed to respond to apparently adequate outpatient treatment. The place of day care as an alternative option in these circumstances has yet to be fully evaluated but seems promising.
122 CONFIDENTIAL Second draft for consultation September 2003
6.5.5 Satisfaction with service setting/ configuration and adherence to treatment The area of patient and carer perceptions has been the focus of some research in the area of eating disorders. Much of this work has been focused on perceptions of inpatient treatment for anorexia nervosa or on mixed or poorly defined populations of people with eating disorders, therefore it should be treated with caution. This area of research is potentially informative when considering the relative merits of different service configurations although the major focus of such work has been on improving the acceptability of services, which may have benefits in terms of improved attendance rates and increased involvement with and effectiveness of programmes and treatments prescribed (Matoff & Matoff, 2001; Swain-Campbell et al., 2001). Taking account of user and carer perceptions when designing and delivering services may also facilitate help seeking over a prolonged period in people with recurrent mental health problems (Buston, 2002). This in turn may contribute to reduced morbidity.
Individuals with eating disorders, and anorexia nervosa in particular, are often described as being ambivalent about seeking treatment. Unlike most other psychiatric conditions, core features of eating disorders can be highly valued by the patient. In addition, the hospital environment can contribute to a sense of passivity and vulnerability, which can be linked to an increased sense of loss of control, one of the central characteristics of an eating disorder (Eivors et al., 2003). Such factors can contribute to a degree of reluctance to engage fully in interventions, resulting in relatively high levels of treatment refusal and premature dropout, with related implications for long term recovery and health-care costs (Kahn & Pike, 2001; Swain-Campbell et al., 2001). Some centres report that up to 50% of patients prematurely cease contact after assessment at an eating disorders service (Button et al., 1997; Vandereycken & Pierloot, 1993). Yager et al. (1989) draw attention to the fact that studies from many eating disorders centres suggest that dropout rates for virtually all types of treatment are considerable, suggesting relatively high levels of patient dissatisfaction with services and treatments offered. Noordenbos et al. (1988) suggest that in people with a long standing eating disorder, patients’ dissatisfaction may be compounded by clinicians’ beliefs about ‘chronic’ or ‘untreatable’ presentations.
People receiving inpatient treatment for anorexia nervosa have been found to be twice as likely to drop out of treatment compared to general psychiatric inpatients (Kahn & Pike, 2001). Reasons for dropout are likely to be varied and complex. One study found that dropout from adult inpatient treatment was modestly predicted by anorexia nervosa subtype (being greater in the binge-purge subtype) and length of illness (with associated higher number of previous hospitalisations) and not to severity of eating disorders symptomatology or associated psychopathology (Kahn & Pike, 2001). Another study concluded that very little of formal treatment is regarded by
123 CONFIDENTIAL Second draft for consultation September 2003
patients as essential to recovery in those who have been treated for and recovered from anorexia nervosa (Maine, 1985). Such findings suggested a complicated relationship between service setting, clinical outcome and patient experience that is difficult to tease out. Models of service provision currently vary considerably. Newton (2001) reports that although various surveys have identified strengths and weaknesses in existing service provision, this information seems to have had little impact on service planning. Assessment of user and carer satisfaction specifically in relation to service setting is rarely carried out. Similarly, patient adherence and drop out, specifically in relation to service setting is not usually investigated (Mahon, 2000). Information about service user and carer views tends to come from qualitative studies of service users, based on structured and semi-structured interviews (with data presented descriptively or, for example, using a type of thematic content analysis such as a grounded theory approach) or questionnaires. The latter are often used to measure characteristics of clinical presentation (rather than service setting), which are then correlated with for example, satisfaction or dropout. In some cases forced choice responses about satisfaction or ratings of perceived helpfulness are used, but again these are rarely related specifically to service configurations. Research in the area of satisfaction surveys is particularly prone to bias.
This issue of satisfaction and adherence to treatment was considered across a number of different types of service setting (e.g., outpatient, inpatient, day- patient, generalist, specialist, self-help; local, distant, etc); and across different age groups (adults, adolescents, children).
The review team conducted a new systematic search for all relevant research. No studies were found that specifically investigated comparisons between types of service. Nevertheless, a number of studies provided useful data (Brinch, Isager & Tolstrup, 1988; Buston, 2002; Button et al., 1997; Button & Warren, 2001; Carnell, 1998; Deeble & Bhat, 1991; Eivors et al., 2003; Haigh & Treasure, 2003; Hsu, Crisp & Callender, 1992; Kahn & Pike, 2000; Kopec- Schrader et al., 1993; Le Grange & Gelman, 1998; Mahon, 2000; Maine, 1985; Malson et al., unpublished; Matoff & Matoff, 2001; Newton, 2001; Newton et al., 1993a; Noordenbos et al., 1988; Pettersen & Rosenvinge, 2002; Rosenvinge & Klusmeier, 2000; Sharkey-Orgnero, 1999; Stockwell et al., 1987; Sturmey, 1992; Swain Campbell, Surgenor & Snell, 2001; Tozzi et al., 2003; Vandereycken & Pierloot, 1983; Yager et al., 1989). Seven further studies were consulted, but not used directly (Bowers & Andersen, 1994; Greenwood et al., 1999; Lemberg & May, 1991; Neiderman et al., 2000; Newton, Hartley & Sturmey, 1993b; Wilhelm & Clarke, 1998; Zipfel et al., 2002).
The following findings emerged from the papers considered in this section:
124 CONFIDENTIAL Second draft for consultation September 2003
• Greater treatment satisfaction is most commonly reported in relation to outpatient treatment (particularly regarding individual and group therapy) (Rosenvinge & Klusmeier, 2000; Newton et al., 1993a)
• Lengthy waiting times for outpatient treatment have been identified as a major reason for being dissatisfied with health care, and problems and delays in accessing medical or other appropriate help are associated with lower levels of satisfaction (Rosenvinge & Klusmeier, 2000; Buston, 2002)
• Carers and parents of adolescents have identified a lack of and need for support, involvement and education about eating disorders for themselves (Kopec-Schrader et al., 1993; Haigh & Treasure, 2003)
• Continuity of care with an individual professional has been found to be valued – this could be taken to support the notion of comprehensive services which can be flexible in terms of intensity (Buston, 2002)
• People who perceive their therapists to be ‘experts on eating disorders’ have been found to be more satisfied with treatment. (‘Expert’ includes concepts of ‘understanding’ and having ‘knowledge of eating disorders’ with the former being rated as the more important therapist attribute) (Rosenvinge & Klusmeier, 2000)
• Informal supportive social relationships outside the treatment setting and a supportive relationship with a partner have been identified as important in recovery in people with eating disorders (Maine, 1985; Pettersen & Rosenvinge, 2002; Tozzi et al., 2003)
• Surveys of self-help groups suggest that they are regarded by the majority of attendees as helpful, being positively associated with social involvement and supportive sharing. However, self –help groups can be experienced by some individuals as upsetting or promoting competitiveness in terms of weight loss and eating disorders behaviours (Deeble & Bhat, 1991)
• Some people who have received treatment for anorexia nervosa report a general negative perception of being treated as an illness or diagnosis rather than as a person (Brinch et al., 1988, Malson et al.). This may be associated with reduced likelihood of seeking psychiatric help later
• Perceived negative staff attitudes, and an overemphasis on weight and weight restoration have been identified as unhelpful or harmful aspects of treatment by adults receiving inpatient treatment (Button & Warren, 2001). An overemphasis on weight gain has also been found to be associated with drop out from treatment (Newton et al., 1993a)
125 CONFIDENTIAL Second draft for consultation September 2003
• The acceptability of inpatient treatment for anorexia nervosa in adolescence has been rated as low in regards to feeling pressured and watched, with authoritarian and restricting aspects of therapy causing anger and ambivalence (Brinch et al., 1988)
• Adolescents have reported a sense of lack of privacy and intrusiveness in the context of inpatient care (Buston, 2002)
• Parents of adolescents with anorexia nervosa have reported feeling blamed for their child’s eating disorder by clinicians providing treatment (Sharkey–Orgnero, 1999).
6.5.5.1 Clinical summary A wide range of views, experiences and levels of satisfaction is expressed, with responses ranging from totally satisfied to totally dissatisfied in relation to specific service settings. No consistent pattern emerges in relation to type of service setting, or age, with different people being satisfied or dissatisfied with different aspects of services or service settings. Treatment adherence, problem severity and eventual outcome are not always related to satisfaction and acceptability (Stockwell et al., 1987; Sturmey, 1992) and service users attribute a wide range of factors (independent of service setting) to recovery (Hsu et al., 1992; LeGrange & Gelman, 1998).
It is common that individuals remain ambivalent about treatment received, particularly those with anorexia nervosa (Brinch et al., 1988, Carnell, 1998). Those who have anorexia nervosa in adolescence appear most likely to recall their treatment (whether inpatient or outpatient) in negative terms. This attitude tends to persist and does not appear to be related to treatment duration or intensity (Brinch et al., 1988; Buston, 2002). The ambivalence characteristic of people with eating disorders in relation to treatment may be an important issue. This ambivalence stems in part from the functional aspects of the disorder itself, and must form part of the backdrop against which views about satisfaction are interpreted. In this respect it might be understandable that some suggest that unlike other psychiatric disorders, service user dissatisfaction will tend to be high in eating disorders (e.g., Swain Campbell et al., 2001).
People (in Norway) asked to list recommendations to increase/improve health care services for people with eating disorders most frequently identified ‘improving clinical competence and knowledge about eating disorders amongst GP’s’ (Rosenvinge & Klusmeier, 2000). The opportunity to talk and be understood seems to be more important than (type of) formalised psychotherapy or service setting in terms of user satisfaction (Button & Warren, 2001).
126 CONFIDENTIAL Second draft for consultation September 2003
Finally, given that self help groups do quite well on very or fairly helpful ratings, the fact that self help group users tend to use other professional services, and the finding that supportive social relationships outside formal treatment settings are associated with recovery, it seems more could be possible in terms of integrated working between statutory and voluntary services/agencies.
6.5.6 Relationship between service setting and risk of death, suicide and self-harm Anorexia nervosa is associated with an increased mortality (Nielsen, 2001). People at risk of death because of their extreme physical state are likely to be considered – appropriately – for admission to hospital. However, there is unlikely to be systematic evidence to support such practice. Furthermore, people suffering from anorexia nervosa may be at increased risk of self-harm and suicide (Favaro & Santonastaso, 2000). This is especially so with respect of the binge-purging sub-type of anorexia nervosa.
6.5.7 Compulsory admission and treatment A decision to compulsorily treat people with eating disorders occurs infrequently but does raise debate amongst professionals working in the field, the individuals they treat and their families. People with anorexia nervosa often differ from others with mental health problems in that the central characteristics of the illness are perceived as functional and valued by the individual. The individual can be perceived as ambivalent about recovery and resistant to intervention. In (eating disorders) treatment an emphasis is placed on developing a collaborative therapeutic relationship with the individual. Under conditions of compulsion this may be more difficult. However individuals who have undergone such treatment report contrasting positive and negative views when asked to comment later.
Treatment in this context refers to inpatient treatment of anorexia nervosa in adults, children and adolescents. However in the case of children and adolescents compulsory treatment can take place on an outpatient basis under parental authority, under the Mental Health Act 1983 and more rarely, with specific Court Orders.
There are no reports of compulsory admission for the specific treatment of bulimia nervosa or binge eating disorder (BED) in the UK.
Compulsory admissions and treatment for adults are carried out using the Mental Health Act (MHA) 1983. The Mental Health Act Commission (Guidance note number 3) provides guidance on the use of the Act in anorexia nervosa. Practice with children and adolescents, varies throughout the country. The issues of the “assessment of competence (in children/
127 CONFIDENTIAL Second draft for consultation September 2003
adolescents) and capacity (in adults)” are complex. Legislation used with this population includes the Mental Health Act 1983 and the Children Act 1989.
Individuals with anorexia nervosa may be ambivalent about their treatment and can experience treatment as compulsory whatever their legal status. Often there are no differences between treatment programmes for those admitted and treated compulsorily and those who are not. Patients with eating disorders are sometimes admitted compulsorily for treatment of comorbid conditions.
For the purpose of this guideline, compulsory admission and treatment will be defined as that carried out using the legal powers available under the Mental Health Act 1983 or the Children Act 1989 or the authority of the court.
A further aim of employing compulsion under the MHA (1983) is to offer the individual the protection that is provided for them in the Act. It is important to remember that compulsory treatment does not equate with “feeding against the will of the person” or “force feeding”. It is helpful to hold in mind the distinction between treatment carried out under the legislation with which the individual complies (for whatever reason) and that which the individual resists.
6.5.7.1 Studies considered The review team conducted a new systematic search for all research relevant to the issue of compulsory admission and treatment. No RCTs or meta- analyses were found. However, a few case control and three cohort studies were included.
6.5.7.2 Evidence There is a lack of research into the outcomes of compulsory admission for anorexia nervosa (Russell, 2001; Watson, 2000). The literature in this area relates mainly to those adults treated within specialist settings or, individual case studies characterised by refusal of treatment in children and adolescents. There is a suggestion in the literature that those compulsorily treated have a poorer outcome but there is insufficient evidence to derive any conclusion from this (Ramsay et al., 1999).
The potential benefits of compulsory admission have been identified as weight gain (Ramsay et al., 1999), saving life (Honig & Bentovim, 1996), opportunities for further treatment and avoiding significant harm (Honig & Bentovim, 1996), improvement in mood and concentration and reduction in symptoms (Maloney & Farrell, 1980), and less starvation induced cognitive impairment (MacDonald, 2002).
The potential risks include; obstacles to the development of a therapeutic relationship (Orbach & Rathner, 1998; Lancely & Travers, 1993), negative physical and psychological effects (Dresser & Boisaubin, 1986), negative
128 CONFIDENTIAL Second draft for consultation September 2003 countertransference , stigma associated with the MHA 1983 affecting the persons future (Hebert et al., 1991), damage to self esteem, escalation of resistance (Rathner, 1998; Fichter, 1995).
The literature provides some guidance on when to employ compulsory treatment (Ramsay et al., 1999). There is considerable guidance available to healthcare professionals and other involved in compulsory admissions which offers advice on the protection to all those involved and guidance on the use of ethical decision making (Manley et al., 2002; Goldner, 1997, Honig & Jaffa, 2000) and the obtaining of consent particularly in the child and adolescent field Practitioners should also be aware of the outcome of reported legal challenges to compulsory treatment in anorexia nervosa, including the use of forcible feeding if necessary which have tended to favour those applying for or defending the right to treat (Dolan, 1999,). Proposed new legislation (the reform of the Mental Health Act 1983) may have a major impact on the issue of consent to treatment.
6.5.7.3 Clinical summary Formal admission for the specific treatment of anorexia nervosa or for immediate life saving is necessary in very rare circumstances, where substantial risk cannot be managed in any other way. Little is known about the outcomes of those who are compulsorily treated; they have a poorer outcome than those who are not compulsorily treated, but this may be due to the initial severity of illness on admission. Treatment under conditions of compulsion very rarely involves feeding in the context of active physical resistance. The successful management of anorexia nervosa in these circumstances requires special skill and expertise in the area of compulsory treatment of people with eating disorders.
Special considerations are required when using compulsory treatment with children and adolescents (in particular the assessment of competence to consent). The issues of consent and treatment refusal are complex in children and adolescents (see DOH guidelines – Reference Guide to Consent for Examination for Treatment, 2002) because of the need to assess competence and involve a third party (parents) in the discussion. When refusal of treatment repeatedly occurs expert legal advice should be sought to establish whether the use of mental health legislation (irrespective of the patient’s age) or an application to the Court under the provision of the Children Act 1989 is the best way to proceed. As both physical and psychological development is compromised in patients with early onset anorexia nervosa, treatment should aim to balance the need for urgent weight restoration alongside the educational and social needs of the young person. It is not recommended to continue treatment relying indefinitely on parental consent. The legal basis under which treatment is being carried out should be clearly recorded for this patient group.
129 CONFIDENTIAL Second draft for consultation September 2003
6.5.8 Clinical practice recommendations
6.5.8.1 Inpatient treatment should be considered for people with anorexia nervosa whose disorder is associated with high or moderate physical risk. (C)
6.5.8.2 Patients with anorexia nervosa requiring inpatient treatment should normally be admitted to a setting that can provide the skilled implementation of refeeding with careful physical monitoring and in combination with psychosocial interventions. (C)*
6.5.8.3 Inpatient treatment or day patient treatment should be considered for people with anorexia nervosa whose disorder has not improved with appropriate outpatient treatment, or for whom there is a significant risk of suicide or severe self-harm. (C)
6.5.8.4 Healthcare professionals without specialist experience, or in situations of uncertainty, should consider seeking advice from an appropriate specialist when considering a formal admission for a patient with anorexia nervosa regardless of the age of the patient. (C)
6.5.8.5 Healthcare professionals managing patients with anorexia nervosa, especially that of the binge purging sub-type, should be aware of the increased risk of self-harm and suicide, particularly at times of transition between services or service settings. (C)
Special considerations for children and adolescents
6.5.8.6 Admission of children and adolescents with anorexia nervosa should be to age-appropriate facilities (with the potential for separate children and adolescent services), which have the capacity to provide appropriate educational and related activities. (C)
6.5.8.7 When a young person with anorexia nervosa refuses treatment, which is deemed essential, consideration should be given to the use of the Mental Health Act 1983 or the right of those with parental responsibility to override the young person’s refusal. (C)
6.5.8.8 Continuing treatment relying indefinitely on parental consent should be avoided. It is recommended that the legal basis under which treatment is being carried out should be recorded in the patient’s case notes in all cases but particularly in the case of children and adolescents. (C)
6.5.8.9 For children and adolescents with anorexia nervosa, where issues of consent to treatment are highlighted, healthcare professionals should consider seeking a second opinion. (C)
130 CONFIDENTIAL Second draft for consultation September 2003
6.5.8.10 If the patient with anorexia nervosa and those with parental responsibility refuse treatment, and treatment is deemed to be essential, legal advice should be sought in order to consider proceedings under the Children Act 1989. (C)
6.6 Predicting the outcome of treatment and recovery from anorexia nervosa It is important to be able to predict how people with anorexia nervosa will respond to treatment. For example, those with a good prognosis may respond to less intensive treatment (i.e., outpatient treatment). Knowledge of the factors associated with outcome may also suggest modifications or alternative approaches for patients who do not respond to first line treatments.
Nielsen et al. (1998) reviewed the mortality rate from published outcome studies of anorexia nervosa and bulimia nervosa. In anorexia nervosa, the association of lower weight at presentation and elevated Standardised Mortality Rate (SMR) was highly significant. Age at presentation was also significant with the highest SMR for those presenting between 20 and 29 years of age. A recent review by Steinhausen (2002) considered 119 studies but did not consider issues relating to study or data quality. A number of other papers reviewed various co-morbid conditions and their relationship to outcome in eating disorders (Holderness et al., 1994; Herzog et al., 1996; Rosenvinge et al., 2000). A wide range of potential predictors of outcome have been studied, often those which are routinely or most easily collected at pre- treatment. Some predictors, such as readiness for change, which may be important, are rarely measured.
For the purposes of the guideline, a literature review was carried out using PsycLIT and MEDLINE to identify relevant studies published on or before January 2003. This augmented the search of clinical trials already performed as part of the review of the effectiveness of treatment interventions. A wide range of studies (cohort and treatment studies) of varying quality were identified. Studies of mixed eating disorder populations were excluded unless separate data for anorexia nervosa and bulimia nervosa were reported. Studies of all ages were included and all variables reported in the studies were included in the analysis. Most of the included studies are of inpatients, which limit the generalisability of the findings to less severe populations. Sample size (n ≥ 50) was selected as the key inclusion criterion as it was not possible to derive robust measures of study quality. One exception was made to the key inclusion criteria; studies with a sample size of less than 50 were included where the follow-up period was 4 years or over. A total of 54 studies contributed to the final analysis with sample sizes ranging from 26-422. Many studies had repeated assessments at follow-up, the longest being 21 years (Lowe et al., 2001).
131 CONFIDENTIAL Second draft for consultation September 2003
The wide variability in method (e.g., different measures of a potential predictor, such as family dysfunction) did not permit a meta-analysis. There was insufficient data to justify analysing data separately for treatments, outcome measures or follow-up length. A detailed analysis of the number of participants leaving a study early, as an outcome measure, was beyond the scope of this review.
Outcome varied considerably across studies (e.g., weight, diagnostic status or Morgan-Russell categories). Some studies also focused on co-morbid disorders as predictors of outcome. For consistency, all findings are expressed in relation to poor outcome.
Based on the number of studies showing a positive result relative to the number showing a negative result for each variable examined, a number of possible predictors emerged (for full details see Appendix 11). That is, in people with anorexia nervosa, a low BMI and a number of indices of physical deterioration prior to treatment are associated with a poorer outcome, as is the bulimic sub-type of anorexia nervosa (in particular vomiting). Previous treatment for anorexia nervosa is also associated with poorer outcome, but findings may be confounded by duration or severity of the disorder. Other pre-treatment predictors identified were personality or interpersonal problems, family disturbance, body image disturbance or dissatisfaction and low desired weight. These findings may indicate areas that need to be addressed within treatment. Age over 20 years at presentation is also associated with poorer outcome.
Post-treatment predictors of poor outcome include other psychiatric disorders, such as mood and personality disorders. This highlights the importance of comprehensive psychiatric and psychological interventions. Inadequate weight gain in treatment, low desired weight, and drive for thinness or continued dieting at post-treatment are also associated with poorer long-term outcome. This highlights the importance of addressing attitudes to food and weight within treatment. Poor social adjustment post- treatment is also associated with poorer long-term outcome indicating the potential needs of patients who do not recover.
6.6.1 Clinical practice recommendations
6.6.1.1 Whenever possible patients should be engaged and treated before reaching severe emaciation. This requires both early identification and intervention. Effective monitoring engagement of patients at severely low weight or with falling weight should be a priority. (C)
132 CONFIDENTIAL Second draft for consultation September 2003
7 Treatment and management of bulimia nervosa
7.1 Introduction Bulimia nervosa was first described in 1979 in a paper titled “Bulimia nervosa: an ominous variant of anorexia nervosa” (Russell, 1979). In this paper Russell described bulimia nervosa as “intractable”. This view is no longer held. Within two years of Russell’s paper a report was published describing a promising psychological treatment for the disorder, a specific form of CBT (Fairburn, 1981), and the following year there were two reports that antidepressant drugs also had a beneficial effect (Pope & Hudson, 1982; Walsh et al., 1982) Since then, bulimia nervosa has been the subject of much research interest. There have been over 60 randomised controlled trials evaluating a range of treatments for bulimia nervosa and their findings are remarkably consistent. Evidence-based treatment is certainly possible.
In this chapter the research on psychological treatments is first reviewed. Then the drug studies are considered, followed by the trials evaluating psychological versus antidepressant drug treatment, and their combination. In addition, there are separate sections on the management of the physical aspects of bulimia nervosa and service level interventions. Finally, there is a section on predicting outcome to treatment and recovery from bulimia nervosa. 7.2 Psychological interventions
7.2.1 Introduction The psychological treatment of bulimia nervosa has been the subject of much research interest. Within two years of the publication of Russell’s seminal paper on bulimia nervosa (Russell, 1979), Fairburn (1981) described promising results with a specific form of cognitive behaviour therapy (CBT-BN) (Fairburn, 1981). The following year another promising psychological treatment was described, a form of exposure with response prevention (ERP) (Rosen & Leitenberg, 1982). By the mid-1980s randomised controlled trials evaluating these and other psychological treatments were beginning to be published (Lacey, 1983; Yates & Sambrailo, 1984; Ordman & Kirchenbaum, 1985; Fairburn et al., 1986; Lee & Rush, 1986 Wilson et al., 1986).
Since then research on the treatment of bulimia nervosa has continued apace. Certain trends are apparent. The studies have become larger, and therefore more statistically powerful (e.g., Agras et al., 2000a), and more sophisticated
133 CONFIDENTIAL Second draft for consultation September 2003 in the questions that they have addressed (e.g., Fairburn et al., 1993; Mitchell et al., 2002). They have included studies of mediators of treatment effects (Wilson et al., 2002) and moderators of treatment outcome (Agras et al., 2000b). Generic research methods have improved too with the use of detailed treatment manuals (e.g., Fairburn, 1993, 1997) and procedures to ensure the treatments are well delivered (e.g., Agras et al., 2000a). Standardised measures of outcome have been employed, the Eating Disorder Examination (EDE) (Cooper & Fairburn, 1987; Fairburn & Cooper, 1993) being widely viewed as the “gold standard”. There has also been interest in not simply identifying effective treatments for bulimia nervosa but also in evaluating ways of disseminating them (e.g., Cooper et al., 1996; Treasure et al., 1996; Palmer et al., 2003).
7.2.2 Current practice There have been no UK-based studies of the treatments that are actually provided for patients with bulimia nervosa. Three studies in the USA suggest that evidence-based treatments are not widely used (Crow et al., 1999; Mussell et al., 2000; Haas & Clopton, 2003). There is no reason to think that the situation is any different in the UK. There are likely to be wide variations in the nature of treatment provided, reflecting the resources available and the training and proclivity of the clinician involved. The findings of two UK based studies suggest that most people with bulimia nervosa are not in treatment (Fairburn & Cooper, 1982; Fairburn et al., 1996).
7.2.3 Psychological treatments versus wait-list or placebo control
7.2.3.1 Treatments reviewed The following treatments were included:
• Cognitive behaviour therapy for bulimia nervosa (CBT-BN)
• CBT + exposure with response prevention (CBT + ERP)
• Focal supportive psychotherapy
• Guided self-help (GSH)
• Pure self-help (PSH)
• Simplified dialectical behaviour therapy (simplified DBT).
The Psychological Topic Group established definitions for each treatment (see Glossary). Two members of the Topic Group assessed each study for eligibility and classified each psychological treatment. Where disagreements arose, they were resolved by discussion.
134 CONFIDENTIAL Second draft for consultation September 2003
7.2.3.2 Studies considered10 The review team used the existing Cochrane review ‘Psychotherapy for bulimia nervosa and binging’ (Hay & Bacaltchuk, 2003) as the starting point for this section.
Fourteen trials used in the existing Cochrane review were excluded because they did not meet the inclusion criteria set by the GDG (BACHAR1999, BOSSERT1989, BULIK1998, CARTER1998, ESPLEN1998, LAESSLE1987, LAESSLE1991, LOEB2000, ORDMAN1985, PETERSON1998, THACKWRAY1993, TREASURE1999, WILFLEY1993, WILSON1986). A further 6 trials were excluded from this section, but included in the next section regarding different psychological treatments (AGRAS2000, COOPER1995, FAIRBURN1986, FAIRBURN1991, WALSH1997, WILSON1991). The 6 remaining trials were included here (AGRAS1989, FREEMAN1988, GRIFFITHS1994, LEE1986, TREASURE1994, WOLF1992). In addition, one trial excluded from the Cochrane review (MITCHELL1990) and 6 trials (LEITENBERG1988, MITCHELL2001A, CARTER2003, WALSHSUBMIT, SUNDGOTBORGEN2002, SAFER2001) found during the search for new evidence, were included. Thus, 13 RCTs comparing a psychological treatment with a wait-list control or placebo group, involving 1029 participants, were included.
Out of the included trials, 10 involved CBT-BN (AGRAS1989, FREEMAN1988, GRIFFITHS1994, LEE1986, LEITENBERG1988, MITCHELL1990, SUNDGOTBORGEN2002, TREASURE1994, WALSHSUBMIT, WOLF1992), three involved CBT + ERP (AGRAS1989, LEITENBERG1988, SUNDGOTBORGEN2002), two involved focal supportive psychotherapy (AGRAS1989, FREEMAN1988), one used GSH (WALSHSUBMIT), three used PSH (CARTER2003, MITCHELL2001A, TREASURE1994), and one involved simplified DBT (SAFER2001).
Full details of the studies included in the guideline and the reasons for excluding studies are given in Appendix 17.
7.2.3.3 Evidence statements11 The level of evidence (I, IIb, IIb, III, IV) is given after each statement (see section 3.4.6 for more information about the classification of evidence).
Effect of treatment on remission from binge eating/ purging
10 Here and elsewhere in the guideline, each study considered for use in the guideline is referred to by a study ID (primary author and date of study publication, except where a study is in press or only submitted for publication, then a date is not used). 11 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
135 CONFIDENTIAL Second draft for consultation September 2003
There is strong evidence suggesting that there is a clinically significant difference between CBT-BN and wait-list control with CBT-BN being superior in terms of remission from binge eating (defined as cessation of binge eating) by the end of treatment (N = 3; n = 136; RR = 0.73; 95% CI, 0.61 to 0.88; NNT = 4; 95% CI, 3 to 9). I
There is limited evidence suggesting that there is a clinically significant difference between simplified DBT and wait-list control with simplified DBT being superior in terms of remission from binge eating/ purging (defined as cessation of binge eating/ purging) by the end of treatment (N = 1; n = 31; RR = 0.75; 95% CI, 0.57 to 1.00; NNT = 4; 95% CI, 3 to 100). I There is limited evidence suggesting that it is unlikely that PSH improves remission from binge eating/ purging by the end of treatment:
• There is evidence suggesting that it is unlikely there is a clinically significant difference between PSH and wait-list control in terms of remission from binge eating (defined as cessation of binge eating) by the end of treatment (N = 2; n = 139; RR = 0.96; 95% CI, 0.85 to 1.09). I
• There is evidence suggesting that it is unlikely there is a clinically significant difference between PSH and wait-list control in terms of remission from purging (defined as cessation of purging) by the end of treatment (N = 2; n = 139; RR = 0.97; 95% CI, 0.87 to 1.07). I
There is insufficient or no evidence to determine whether CBT + ERP, focal supportive psychotherapy, or GSH have any impact on remission rates.
Effect of treatment on frequency of binge eating and purging symptoms There is strong evidence that CBT-BN reduces the frequency of binge eating and purging symptoms by the end of treatment:
• There is strong evidence suggesting that there is a clinically significant difference between CBT-BN and wait-list control with CBT-BN being superior on mean frequency of binge eating by the end of treatment (N = 5; n = 185; SMD = -0.75; 95% CI, -1.05 to -0.44). I
• There is strong evidence suggesting that there is a clinically significant difference between CBT-BN and wait-list control with CBT-BN being superior on mean frequency of purging by the end of treatment (N = 6; n = 192; Random Effects SMD = -1.00; 95% CI, -1.63 to -0.36). I
There is strong evidence suggesting that there is a clinically significant difference between CBT + ERP and wait-list control with CBT + ERP being superior on mean frequency of purging by the end of treatment (N = 2; n = 57; SMD = -0.83; 95% CI, -1.37 to -0.28). I
136 CONFIDENTIAL Second draft for consultation September 2003
There is strong evidence that focal supportive psychotherapy reduces the frequency of binge eating and purging symptoms by the end of treatment:
• There is strong evidence suggesting that there is a clinically significant difference between focal supportive psychotherapy and wait-list control with focal supportive psychotherapy being superior on mean frequency of binge eating by the end of treatment (N = 1; n = 50; SMD = -1.12; 95% CI, -1.73 to -0.51). I
• There is strong evidence suggesting that there is a clinically significant difference between focal supportive psychotherapy and wait-list control with supportive psychotherapy being superior on mean frequency of purging by the end of treatment (N = 2; n = 84; SMD = - 1.43; 95% CI, -1.93 to -0.94). I
There is insufficient or no evidence to determine whether GSH, PSH, or simplified DBT have any impact on frequency of binge eating and purging symptoms.
Other effects of treatment There is strong evidence suggesting that there is a clinically significant difference between CBT-BN and wait-list control with CBT-BN being superior with regard to the mean depression score by the end of treatment (N = 3; n = 87; Random Effects SMD = -1.19; 95% CI, -1.99 to -0.39). I
There is strong evidence suggesting that there is a clinically significant difference between CBT + ERP and wait-list control with CBT + ERP being superior with regard to the mean depression score by the end of treatment (N = 1; n = 34 SMD = -1.20; 95% CI, -1.94 to -0.46). I
There is insufficient or no evidence to determine whether focal supportive psychotherapy, GSH, PSH, or simplified DBT have any impact on depression. I
There is insufficient or no evidence to determine whether CBT-BN, CBT + ERP, focal supportive psychotherapy, GSH, PSH, or simplified DBT have any impact on interpersonal/ psychosocial functioning. I
Attrition from the study There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and wait-list control in terms of the number of people leaving the study early due to any reason by end of treatment (N = 9; n = 384; RR = 1.14; 95% CI, 0.74 to 1.74). I
There is insufficient or no evidence to determine whether there is any difference between CBT + ERP, focal supportive psychotherapy, GSH, PSH,
137 CONFIDENTIAL Second draft for consultation September 2003
or simplified DBT and a wait-list control in terms of the number of people leaving the study early due to any reason by end of treatment.
7.2.4 Psychological treatments versus other psychological treatments
7.2.4.1 Treatments reviewed12 The following treatments were included: • Behaviour therapy (BT)
• Cognitive behaviour therapy for bulimia nervosa (CBT-BN)
• CBT + exposure with response prevention (CBT + ERP)
• Focal supportive psychotherapy
• Group CBT
• Guided self-help (GSH)
• Individual CBT
• Interpersonal psychotherapy for bulimia nervosa (IPT-BN)
• Dietary counselling
• Psychodynamic psychotherapy
7.2.4.2 Studies considered The review team used the existing Cochrane review ‘Psychotherapy for bulimia nervosa and binging’ (Hay & Bacaltchuk, 2003) as the starting point for this section.
Fourteen trials used in the existing Cochrane review were excluded because they did not meet the inclusion criteria set by the GDG (BACHAR1999, BOSSERT1989, BULIK1998, CARTER1998, ESPLEN1998, LAESSLE1987, LAESSLE1991, LOEB2000, ORDMAN1985, PETERSON1998, THACKWRAY1993, TREASURE1999, WILFLEY1993, WILSON1986). A further two trials were excluded from this section, but included in the earlier section of psychological treatments versus wait-list control or placebo (GRIFFITHS1994, LEE1986). The 10 remaining trials were included in the
12 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
138 CONFIDENTIAL Second draft for consultation September 2003
present section (AGRAS1989, AGRAS2000, COOPER1995, FAIRBURN1986, FAIRBURN1991, FREEMAN1988, TREASURE1994, WALSH1997, WILSON1991, WOLF1992). In addition, eight trials (BAILERINPRESS, CHEN2003, GARNER1993, HSU2001, JANSEN2002, KIRKLEY1985, LEITENBERG1988, SUNGOTBORGEN2002) found during the search for new evidence, were included. Thus, 18 RCTs comparing two different psychological treatments, involving 1343 participants, were included in this section.
Out of the included trials, three involved a comparison of CBT-BN with BT (FAIRBURN1991, FREEMAN1988, WOLF1992), with follow-up periods ranging from 3 to 12 months. Four trials involved a comparison of CBT with CBT + ERP (BULIK1998, COOPER1995, LEITENBERG1988, WILSON1991) with all using 12 months follow-up, except LEITENBERG1988 (6 months follow-up). Two trials involved a comparison of CBT-BN with IPT-BN (AGRAS2000, FAIRBURN1991) using a follow-up period of 8 and 12 months, respectively. One trial involved a comparison of CBT-BN with psychodynamic psychotherapy (GARNER1993), but presented no follow-up data. Four trials had a comparison of CBT-BN with focal supportive psychotherapy (AGRAS1989, FAIRBURN1986, FREEMAN1988, KIRKLEY1985) using follow-up periods ranging between 3 and 12 months. One trial compared CBT-BN with dietary counselling (SUNGOTBORGEN2002) using a follow-up of 18 months, and one trial compared CBT-BN with GSH (BAILERINPRESS) using a follow-up of 12 months. Another trial compared CBT-BN with PSH (TREASURE1994) with no follow-up.
In addition, one trial compared CBT + ERP with focal supportive psychotherapy (AGRAS1989) and two compared CBT + ERP with dietary counselling (HSU2001, JANSEN2002), all without follow-up data. One trial compared BT with IPT-BN (FAIRBURN1991) using a follow-up period of 12 months, and one compared BT with focal supportive psychotherapy (FREEMAN1988). Finally, one compared individual CBT with group CBT (CHEN2003) with 6 months follow-up.
Full details of the studies included in the guideline and the reasons for excluding studies are given in appendix 17.
7.2.4.3 Evidence statements13
Effect of treatment on remission from binge eating and purging
13 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
139 CONFIDENTIAL Second draft for consultation September 2003
There is evidence from two trials that CBT-BN when compared to IPT-BN improves remission from binge eating and purging by the end of treatment, but is no longer superior at post-treatment follow-up:
• There is strong evidence suggesting that there is a clinically significant difference between CBT-BN and IPT-BN with CBT-BN being superior in terms of remission from binge eating (defined as cessation of binge eating) by the end of treatment (N = 2; n = 270; RR = 0.77; 95% CI, 0.67 to 0.87; NNT = 5; 95% CI, 4 to 20). I
• There is strong evidence suggesting that there is a clinically significant difference between CBT-BN and IPT-BN with CBT-BN being superior in terms of remission from purging (defined as cessation of purging) by the end of treatment (N = 1; n = 220; RR = 0.76; 95% CI, 0.67 to 0.86; NNT = 5; 95% CI, 4 to 8). I
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and IPT-BN in terms of remission from binge eating (defined as cessation of binge eating/ purging) at follow-up (N = 2; n = 270; RR = 0.93; 95% CI, 0.82 to 1.06). I
There is insufficient or no evidence to determine whether there is any difference between CBT-BN and BT on remission rates. I
Effect of treatment on frequency of binge eating and purging There is limited evidence suggesting that there is a clinically significant difference between CBT-BN and CBT + ERP with CBT-BN being superior in terms of mean frequency of binge eating at follow-up (N = 1; n = 25; SMD = - 0.90; 95% CI, -1.73 to -0.07). I
There is limited/ strong evidence that CBT-BN when compared to dietary counselling is superior at reducing the frequency of purging by both the end of treatment and post-treatment follow-up:
• There is limited evidence suggesting that there is a clinically significant difference between CBT-BN and dietary counselling with CBT-BN being superior on mean frequency of purging by the end of treatment (N = 1; n = 31; SMD = -0.95; 95% CI, -1.70 to -0.20). I
• There is strong evidence suggesting that there is a clinically significant difference between CBT-BN and dietary counselling with CBT-BN being superior on mean frequency of purging at follow-up (N = 1; n = 31; SMD = -1.34; 95% CI, -2.13 to -0.55). I
It is unlikely that CBT-BN is superior to BT at reducing the frequency of binge eating and purging by the end of treatment:
140 CONFIDENTIAL Second draft for consultation September 2003
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and BT on mean frequency of binge eating by the end of treatment (N = 3; n = 131; SMD = -0.11; 95% CI, -0.45 to 0.24). I
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and BT on mean frequency of purging by the end of treatment (N = 3; n = 131; SMD = 0.08; 95% CI, - 0.27 to 0.42). I
It is unlikely that CBT-BN is superior to IPT-BN at reducing the frequency of binge eating by both the end of treatment or post-treatment follow-up:
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and IPT-BN on mean frequency of binge eating by the end of treatment (N = 2; n = 262; SMD = -0.24; 95% CI, -0.48 to 0.01). I
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and IPT-BN in terms of mean frequency of binge eating at follow-up (N = 2; n = 257; SMD = -0.04; 95% CI, -0.29 to 0.20). I
It is unlikely that CBT-BN is superior to focal supportive psychotherapy at reducing the frequency of binge eating and purging by the end of treatment:
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and focal supportive psychotherapy on mean frequency of binge eating by the end of treatment (N = 3; n = 111; SMD = 0.00; 95% CI, -0.37 to 0.38). I
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and focal supportive psychotherapy on mean frequency of purging by the end of treatment (N = 4; n = 144; SMD = -0.13; 95% CI, -0.46 to 0.20). I
It is unlikely that CBT-BN is superior to PSH at reducing the frequency of binge eating by the end of treatment:
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BN and PSH on mean frequency of binge eating by the end of treatment (N = 1; n = 80; SMD = 0.03; 95% CI, -0.43 to 0.49). I
141 CONFIDENTIAL Second draft for consultation September 2003
There is limited evidence from one trial that GSH when compared to CBT-BN is superior at reducing the frequency of binge eating and purging by the end of treatment:
• There is limited evidence suggesting that there is a clinically significant difference between CBT-BN and GSH with GSH being superior on mean frequency of binge eating by the end of treatment (N = 1; n = 56; SMD = 1.20; 95% CI, 0.63 to 1.78). I
• There is limited evidence suggesting that there is a clinically significant difference between CBT-BN and GSH with GSH being superior on mean frequency of purging by the end of treatment (N = 1; n = 56; SMD = 0.55; 95% CI, 0.01 to 1.08). I
Other effects of treatment There is insufficient or no evidence to determine whether CBT-BN differs from BT in its effects on depression, general psychiatric symptoms, or psychosocial/ interpersonal functioning by the end of treatment or post- treatment follow-up. I
There is insufficient or no evidence to determine whether CBT-BN differs from CBT + ERP in its effects on depression, general psychiatric symptoms, or psychosocial/ interpersonal functioning by the end of treatment or post- treatment follow-up. I
There is insufficient or no evidence to determine whether CBT-BN differs from IPT-BN in its effects on depression or psychosocial/ interpersonal functioning by the end of treatment or post-treatment follow-up. I
There is insufficient or no evidence to determine whether CBT-BN differs from psychodynamic psychotherapy in its effects on psychosocial/ interpersonal functioning by the end of treatment or post-treatment follow- up, or on general psychiatric symptoms at follow-up. I
There is insufficient evidence to determine whether CBT-BN differs from focal supportive psychotherapy in its effects on general psychiatric symptoms or psychosocial/ interpersonal functioning by the end of treatment or post- treatment follow-up. I
There is no evidence to determine whether CBT-BN differs from dietary counselling in its effects on depression, general psychiatric symptoms or psychosocial/ interpersonal functioning by the end of treatment or post- treatment follow-up. I
142 CONFIDENTIAL Second draft for consultation September 2003
There is insufficient or no evidence to determine whether CBT-BN differs from GSH or PSH in its effects on depression, general psychiatric symptoms or psychosocial/ interpersonal functioning by the end of treatment or post- treatment follow-up. I
There is insufficient or no evidence to determine whether CBT + ERP differs from focal supportive psychotherapy or dietary counselling in its effects on depression, general psychiatric symptoms or psychosocial/ interpersonal functioning by the end of treatment or post-treatment follow-up. I
There is insufficient or no evidence to determine whether BT differs from IPT- BN or focal supportive psychotherapy in its effects on depression, general psychiatric symptoms or psychosocial/ interpersonal functioning by the end of treatment or post-treatment follow-up. I
Acceptability of treatment There is insufficient evidence to determine whether CBT-BN is more, or less, acceptable to people with bulimia nervosa when compared with BT, CBT + ERP, IPT-BN, psychodynamic psychotherapy, focal supportive psychotherapy, dietary counselling, GSH or PSH. I
There is insufficient evidence to determine whether CBT + ERP is more, or less, acceptable to people with bulimia nervosa when compared with focal supportive psychotherapy or dietary counselling. I
There is insufficient evidence to determine whether BT is more, or less, acceptable to people with bulimia nervosa when compared with IPT-BN or focal supportive psychotherapy. I
7.2.5 Special considerations in the management of children and adolescents Bulimia nervosa is rarely, if ever, seen in children. It does occur in adolescents although in clinical practice most patients are young adults. There has been no research on the treatment of adolescents with bulimia nervosa. This omission needs to be rectified. In line with much current clinical practice, the GDG took the view that, subject to adaptation to their age, circumstances and level of development, adolescent patients with bulimia nervosa should receive the same type of treatment as adults with the disorder. In the treatment of adolescents with CBT-BN consideration should also be given to the appropriate involvement of the family.
7.2.6 Clinical summary Four main conclusions may be drawn from these analyses and the studies upon which they are based. First, until recently, most of the studies of the psychological treatment of bulimia nervosa have been small in size and therefore lacking in statistical power. They have therefore been vulnerable to
143 CONFIDENTIAL Second draft for consultation September 2003
Type II error. Thus, the great majority of the many statistically non-significant findings cannot be interpreted. Second, the weight of evidence as measured in terms of the strength and consistency of the findings and the number of relevant studies indicates that CBT-BN (delivered on a one-to-one basis) is the most effective treatment for bulimia nervosa. Third, CBT+ ERP is also effective in the short-term. However, its use has fallen from favour because it is difficult to implement and disliked by patients (Bulik et al., 1998) and time consuming. Lastly, IPT-BN appears to be as effective as CBT-BN at 8 to 12- month post-treatment follow-up (i.e., one-year after starting treatment). Prior to this, CBT-BN is more effective than IPT-BN.
7.2.7 Clinical practice recommendations
7.2.7.1 As a possible first step, patients with bulimia nervosa should be encouraged to follow an evidence-based self-help programme. (B)
7.2.7.2 Healthcare professionals should consider providing direct encouragement and support to patients undertaking an evidence- based self-help programme as this may improve outcomes. This may be sufficient treatment for a limited subset of patients. (B)
7.2.7.3 Cognitive behaviour therapy for bulimia nervosa (CBT-BN), a specifically adapted form of CBT, should be offered to most adults with bulimia nervosa. (A)*
7.2.7.4 An adequate course of CBT-BN should normally be of 16 to 20 sessions over 4 to 5 months. (A)
7.2.7.5 Older adolescents with bulimia nervosa may be treated with CBT-BN adapted as needed to suit their age, circumstances and level of development. (C)*
7.2.7.6 When people with bulimia nervosa have not responded to or do not want CBT, other psychological treatments should be considered. (B)
7.2.7.7 Interpersonal psychotherapy should be considered as an alternative to CBT, but patients should be informed it takes 8 - 12 months to achieve results comparable to CBT. (B)
7.3 Pharmacological interventions
7.3.1 Introduction A diverse network of neurotransmitters/neurohormones are involved in the central and peripheral control of appetite and satiety. A wide array of drugs
144 CONFIDENTIAL Second draft for consultation September 2003
which act on various receptors within these pathways have been examined in the treatment of bulimia nervosa.
7.3.2 Current practice Antidepressants are often employed as a first line treatment as they are easily used primary care.
Drugs are not as acceptable or as well tolerated as psychological treatments in this patient group. Only short-term effects have been studied and the outcome measures used are often not comparable to those used in studies of psychological treatments.
People with bulimia nervosa have an increased risk of self-harm and so risks of overdose need to be considered. Drugs that require dietary restrictions, such as monoamine-oxidase inhibitors (MAOIs) may be less appropriate in this group. They may also be using a wide variety of non-prescription medication the effects of which are unknown and may interact adversely with prescription medication. Particular consideration needs to be given to the possibility of pregnancy and breast-feeding. Very few drugs are recommended for children and adolescents aged less than 18. There are safety data available for other conditions for sertraline and amisulpride in the under 18 group but the use of these medications has not been studied in adults with bulimia nervosa.
7.3.3 Antidepressant drug treatment In clinical practice, it has often been found that any drug effect is poorly sustained and it is often necessary to switch medication in an effort to sustain a remission of symptoms. The effective dose of fluoxetine is 60 mg, and as such higher than the standard dose for depression. No other drug studies have compared different doses.
7.3.3.1 Drugs reviewed The following drugs were included: • Antidepressants
o MAOIs (moclobemide, phenelzine)
o SSRIs (Fluoxetine)
o Tricyclic antidepressants (desipramine, imipramine)
o Other antidepressants (bupropion, trazodone, mianserin)
• Opioid antagonist (naltrexone)
• Antiemetics.
145 CONFIDENTIAL Second draft for consultation September 2003
Drugs that have had their licences withdrawn from the UK were not included in the guideline. Although there are studies assessing the use of lithium carbonate, these were not included, as it was felt inadvisable to use this drug given the potential risk of toxicity in people with bulimia nervosa.
7.3.3.2 Studies considered The review team used the existing Cochrane review ‘Antidepressants versus placebo for people with bulimia nervosa’ (Bacaltchuk & Hay, 2003) as the starting point for the present analysis.
Five trials used in the existing Cochrane review were excluded because they did not meet the inclusion criteria set by the GDG (AGRAS1987, KENNEDY1988, KENNEDY1993, MITCHELL1984, ROTHSCHILD1994). The 11 remaining trials were included in this section (FLUOXETINEBULIM1992, GOLDSTEIN1995, HORNE1988, KANERVA1994, MCCANN1990, MITCHELL1990, POPE1983, POPE1989, SABINE1983, WALSH1987, WALSH1991). In addition, four trials of antidepressants (CARRUBA2001, MITCHELL2001A, RAMANO2002, WALSHSUBMIT), two trials of naltrexone (HUSEMAN1990, MITCHELL1989A) and two of ondansetron (FARIS1989, FARIS2000) were found during the search for new evidence. Thus, 19 RCTs involving 1851 participants were included.
Of the 19 trials, 15 compared antidepressants with placebo. Of these, four compared tricyclic antidepressants with placebo (MCCANN1990, MITCHELL1990, POPE1983, WALSH1991), five compared SSRIs with placebo (FLUOXETINEBULIM1992, GOLDSTEIN1995, KANERVA1994, MITCHELL2001A, WALSHSUBMIT), two compared MAOIs with placebo (WALSH1987, CARRUBA2001), and three compared other antidepressants with placebo (HORNE1988, POPE1989, SABINE1983). The trials ranged in duration from 26 to 365 days, with a median of 56 days.
Full details of the studies included in the guideline and the reasons for excluding studies are given in appendix 17.
7.3.3.3 Evidence statements14 The data were analysed in two ways. First, by combining the results across all classes of antidepressants (referred to below as ‘antidepressants’). Secondly, by combining the results across each class separately (i.e., MAOIs, SSRIs, and tricyclics).
Effect of treatment on remission from binge eating/ purging
14 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
146 CONFIDENTIAL Second draft for consultation September 2003
There is limited evidence suggesting that there is a clinically significant difference between antidepressants and placebo, with antidepressants being superior in terms of remission from binge eating/ purging (defined as cessation of binge eating/ purging) by the end of treatment (N = 6; n = 697; RR = 0.88; 95% CI, 0.83 to 0.94; NNT = 9; 95% CI, 6 to 15). I
There is limited evidence suggesting that there is a clinically significant difference between the MAOIs and placebo with MAOIs being superior in terms of remission from binge eating (defined as cessation of binge eating) by the end of treatment (N = 1; n = 62; RR = 0.77; 95% CI, 0.62 to 0.95; NNT = 5; 95% CI, 3 to 17). I
There is insufficient evidence to determine whether SSRIs or tricyclics are preferable in improving remission rates by the end of treatment. I
Effect of treatment on frequency of binge eating and purging There is strong evidence suggesting that there is a clinically significant difference between antidepressants and placebo with antidepressants being superior in terms of clinical improvement (defined as at least a 50% reduction in the frequency of binge eating) by the end of treatment (N = 6; n = 855; RR = 0.68; 95% CI, 0.60 to 0.78; NNT = 5; 95% CI, 4 to 8). I
However, using a continuous measure of binge eating or purging, there is insufficient evidence to determine whether antidepressants reduce the frequency of binge eating and purging by the end of treatment. I
There is strong evidence from three trials that SSRIs when compared to placebo produce clinical improvement in binge eating and purging by the end of treatment:
• There is strong evidence suggesting that there is a clinically significant difference between the SSRIs and placebo with SSRIs being superior in terms of clinical improvement (defined as at least a 50% reduction in frequency of binge eating) by the end of treatment (N = 3; n = 706; RR = 0.73; 95% CI, 0.62 to 0.84; NNT = 6; 95% CI, 5 to 12). I
• There is strong evidence suggesting that there is a clinically significant difference between SSRIs and placebo with SSRIs being superior in terms of clinical improvement (defined as at least a 50% reduction in the frequency of purging) by the end of treatment (N = 2; n = 656; RR = 0.66; 95% CI, 0.57 to 0.76; NNT = 5; 95% CI, 4 to 7). I
There is strong evidence suggesting that there is a clinically significant difference between tricyclics and placebo with tricyclics being superior in terms of clinical improvement (defined as at least a 50% reduction in the frequency of binge eating) by the end of treatment (N = 1; n = 22; RR = 0.30; 95% CI, 0.11 to 0.80; NNT = 2; 95% CI, 2 to 4). I
147 CONFIDENTIAL Second draft for consultation September 2003
There is strong evidence suggesting that there is a clinically significant difference between tricyclics and placebo with tricyclics being superior on mean frequency of binge eating by the end of treatment (N = 3; n = 120; SMD = -0.82; 95% CI, -1.20 to -0.45). I
There is insufficient evidence to determine whether appetite suppressants or antiemetics reduce the frequency of binge eating or purging when compared to placebo by the end of treatment. I
Other effects of treatment There is limited evidence suggesting that there is a clinically significant difference between antidepressants and placebo, with antidepressants being superior in terms of mean depression scores by the end of treatment (N = 6; n = 293; SMD = -0.28; 95% CI, -0.51 to -0.05). I
There is limited evidence suggesting that there is a clinically significant difference between tricyclics and placebo with tricyclics being superior in terms of mean depression scores by the end of treatment (N = 3; n = 120; SMD = -0.47; 95% CI, -0.83 to -0.10). I
There is insufficient evidence to determine whether MAOIs or SSRIs reduce the symptoms of depression by the end of treatment. I
Acceptability of treatment There is limited evidence from five trials favouring SSRIs over placebo in terms of treatment acceptability:
• There is limited evidence suggesting that there is a clinically significant difference between the SSRIs and placebo with SSRIs being superior in terms of the number of people leaving the study early due to any reason by end of treatment (N = 5; n = 803; RR = 0.79; 95% CI, 0.67 to 0.95). I
There is strong evidence from four trials that tricyclics are less acceptable to people with bulimia nervosa than placebo by the end of treatment:
• There is strong evidence suggesting that there is a clinically significant difference between tricyclics and placebo with placebo being superior in terms of the number of people leaving the study early due to any reason by end of treatment (N = 4; n = 217; RR = 2.03; 95% CI, 1.18 to 3.49; NNH = 7; 95% CI, 4 to 20). I
There is insufficient evidence to determine whether appetite suppressants or antiemetics are more, or less, acceptable to people with bulimia nervosa than placebo. I
148 CONFIDENTIAL Second draft for consultation September 2003
Tolerability of treatment There is limited evidence suggesting that there is a clinically significant difference between antidepressants and placebo with placebo being superior in terms of the number of people leaving the study early due to adverse events by end of treatment (N = 9; n = 1078; RR = 1.90; 95% CI, 1.20 to 2.99; NNH = 20; 95% CI, 13 to 50). I
There is insufficient evidence to determine whether a specific class of antidepressant is more, or less, tolerated in people with bulimia nervosa. I
7.3.4 Special considerations in the management of children and adolescents Antidepressant drugs may be used for the treatment of bulimia nervosa in adolescents but they are not licensed for this age group and there is no evidence base for this practice. They should not be considered as a first line treatment in adolescent bulimia nervosa.
7.3.5 Clinical summary There are a limited number of studies and inevitably this must lead to rather tentative conclusions. Antidepressants, in particular, fluoxetine can produce benefits in overall symptomatology but the evidence is not as strong for specific behaviours such as binge eating and purging. Antidepressants, specifically SSRIs, appear to be reasonably tolerated. The major problem in interpreting the outcome of these trials is their short duration and lack of follow-up data for most trials and the consequent absence of any evidence for long-term effectiveness.
7.3.6 Clinical practice recommendations
7.3.6.1 As an alternative or additional first step to using an evidence-based self-help programme, adults with bulimia nervosa may be offered a trial of an antidepressant drug. (B)*
7.3.6.2 Patients should be informed that antidepressant drugs can reduce the frequency of binge eating and purging, but the long-term effects are unknown. Any beneficial effects will be rapidly apparent. (B)
7.3.6.3 Selective serotonin reuptake inhibitors (SSRIs) (specifically fluoxetine) are the drugs of first choice for the treatment of bulimia nervosa in terms acceptability, tolerability and reduction of symptoms. (C)
7.3.6.4 For people with bulimia nervosa, the effective dose of fluoxetine is higher than for depression (60 mg daily). (C)
7.3.6.5 No drugs, other than antidepressants, are recommended for the treatment of bulimia nervosa. (B)
149 CONFIDENTIAL Second draft for consultation September 2003
7.4 Antidepressant drugs compared to psychological interventions and their combination
7.4.1 Introduction Psychological and pharmacological treatments for bulimia nervosa have not only been evaluated on their own but they have also been compared. In addition, they have been studied in combination to see whether any advantages come from using them together. The only drugs to be studied in this way have been antidepressant drugs. These studies are now reviewed starting with the direct comparative studies and then the evaluations of the two treatments combined.
7.4.2 Current practice There are no data available on the combined use of antidepressant drugs and psychological treatment in the management of bulimia nervosa in routine clinical practice.
7.4.3 Psychological treatments versus antidepressant drugs
7.4.3.1. Treatments reviewed The following psychological treatments were included: • CBT
• Focal supportive psychotherapy
• PSH.
The following antidepressant drugs were included: • SSRIs (fluoxetine)
• Tricyclic antidepressants (desipramine, imipramine).
7.4.3.2. Studies considered The review team used the existing Cochrane review ‘Antidepressants versus psychological treatments and their combination for bulimia nervosa’ (Bacaltchuk, Hay & Trefiglio, 2003) as the starting point for the present analysis.
All five trials comparing a psychological treatment with an antidepressant, used in the existing Cochrane review, were included in the present section (AGRAS1992, GOLDBLOOM1997, MITCHELL1990, LEITENBERG1994, WALSH1997). In addition, three trials (MITCHELL2001A, JACOBI2002, WALSHSUBMIT) found during the search for new evidence, were included. Thus, eight RCTs involving 694 participants were included.
150 CONFIDENTIAL Second draft for consultation September 2003
Out of the included trials, five compared CBT with an antidepressant (AGRAS1992, GOLDBLOOM1997, JACOBI2002, LEITENBERG1994, WALSH1997), and two compared PSH with an antidepressant (MITCHELL2001A, WALSHSUBMIT). In addition, WALSH1997 compared focal supportive psychotherapy with an antidepressant.
Full details of the studies included in the guideline and the reasons for excluding studies are given in Appendix 17.
7.4.3.3. Evidence statements15 The data were analysed by comparing each psychological treatment to any antidepressant, irrespective of drug class.
Effect of treatment on remission from binge eating/ purging There is limited evidence from five trials favouring CBT over antidepressants with regard to remission from binge eating and purging (defined as cessation of binge eating/ purging) by the end of treatment:
• There is limited evidence suggesting that there is a clinically significant difference between CBT and antidepressants with CBT being superior in terms of remission from binge eating by the end of treatment (N = 5; n = 270; RR = 0.78; 95% CI, 0.67 to 0.92; NNT = 6; 95% CI, 4 to 15). I
• There is limited evidence suggesting that there is a clinically significant difference between CBT and antidepressants with CBT being superior in terms of remission from purging by the end of treatment (N = 5; n = 196; RR = 0.78; 95% CI, 0.66 to 0.92; NNT = 6; 95% CI, 4 to 15). I
There is insufficient evidence to determine whether CBT differs from antidepressants with respect to remission from binge eating or purging at post-treatment follow-up. I
There is insufficient or no evidence to determine whether PSH or focal supportive psychotherapy differ from antidepressants with respect to remission from binge eating or purging by the end of treatment or post- treatment follow-up. I
Effect of treatment on frequency of binge eating and purging Overall, it is unlikely that there is any difference between CBT and antidepressants with respect to the frequency of binge eating or purging by the end of treatment:
15 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
151 CONFIDENTIAL Second draft for consultation September 2003
• There is insufficient evidence to determine if there is a clinically significant difference between CBT and antidepressants on mean frequency of binge eating by the end of treatment (N = 4; n = 149; SMD = -0.19; 95% CI, -0.52 to 0.14). I
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT and antidepressants on mean frequency of purging by the end of treatment (N = 5; n = 158; SMD = - 0.14; 95% CI, -0.46 to 0.19). I
There is insufficient evidence to determine whether CBT differs from antidepressants with respect to the frequency of binge eating or purging at post-treatment follow-up. I
There is insufficient or no evidence to determine whether PSH or focal supportive psychotherapy differ from antidepressants with respect to the frequency of binge eating or purging by the end of treatment or post- treatment follow-up. I
Other effects of treatment There is insufficient or no evidence to determine whether CBT, PSH or focal supportive psychotherapy differ from antidepressants on depression or general psychiatric symptoms by the end of treatment or post-treatment follow-up. I
Acceptability of treatment There is insufficient evidence to determine whether CBTis more, or less, acceptable to people with bulimia nervosa when compared with antidepressants. I
However, closer inspection of the data indicated significant heterogeneity in the data set. A sensitivity analysis was conducted by removing one study (JACOBI2002) from the analysis as this study appeared to be an outlier. The authors of this study speculated that their findings might differ from previous studies due to cultural differences in the health care systems (i.e., the study was conducted in Germany, whereas the other studies were North American). The sensitivity analysis indicated that:
• There is strong evidence suggesting that there is a clinically significant difference favouring CBT-BN over antidepressants in terms of the number of people leaving the study early due to any reason by end of treatment (N = 3; n = 149; RR = 0.48; 95% CI, 0.28 to 0.83). I
There is insufficient or no evidence to determine whether PSH or focal supportive psychotherapy are more, or less, acceptable to people with bulimia nervosa when compared with antidepressants. I
152 CONFIDENTIAL Second draft for consultation September 2003
Tolerability of treatment There is insufficient or no evidence to determine whether antidepressants have more, or less, side effects in people with bulimia nervosa when compared with CBT, PSH or focal supportive psychotherapy. I
7.4.4 Psychological treatments versus the combination of psychological treatment and antidepressants
7.4.4.1 Treatments reviewed The following psychological treatments were included: • CBT
• Focal supportive psychotherapy
• Dietary counselling
• PSH.
The following antidepressant drugs were included: • SSRIs (fluoxetine)
• Tricyclic antidepressants (desipramine, imipramine).
7.4.4.2 Studies considered The review team used the existing Cochrane review ‘Antidepressants versus psychological treatments and their combination for bulimia nervosa’ (Bacaltchuk, Hay & Trefiglio, 2003) as the starting point for the present analysis.
All seven trials comparing a psychological treatment to the combination of a psychological treatment and an antidepressant, used in the existing Cochrane review, were included in this section (AGRAS1992, BEUMONT1997, FICHTER1991, GOLDBLOOM1997, LEITENBERG1994, MITCHELL1990, WALSH1997). In addition, three trials (MITCHELL2001A, JACOBI2002, WALSHSUBMIT) found during the search for new evidence, were included here. Thus, 10 RCTs involving 801 participants were included.
Out of the included trials, seven used CBT as the psychological treatment (AGRAS1992, FICHTER1991, GOLDBLOOM1997, JACOBI2002, LEITENBERG1994, MITCHELL1990, WALSH1997), two used PSH (MITCHELL2001A, WALSHSUBMIT) and one used dietary counselling (BEUMONT1997). In addition, WALSH1997 included a focal supportive psychotherapy group.
153 CONFIDENTIAL Second draft for consultation September 2003
Full details of the studies included in the guideline and the reasons for excluding studies are given in appendix 17.
7.4.4.3 Evidence statements16 The data were analysed by comparing each psychological treatment to the combination of the psychological treatment and any antidepressant, irrespective of drug class.
Effect of treatment on remission from binge eating/ purging There is insufficient evidence to determine whether CBT differs from the combination of CBT and antidepressants with respect to remission from binge eating or purging by either the end of treatment or post-treatment follow-up. I
There is insufficient evidence to determine whether PSH or focal supportive psychotherapy differs from the combination of PSH or focal supportive psychotherapy and antidepressants with respect to remission from binge eating or purging by either the end of treatment or post-treatment follow-up. I
There is insufficient evidence to determine whether dietary counselling differs from the combination of dietary counselling and antidepressants with respect to remission from binge eating or purging by the end of treatment. However, at follow-up there is evidence from one study favouring the combination:
• There is limited evidence to suggest that there is a clinically significant difference between dietary counselling and a combination of dietary counselling and an antidepressant with dietary counselling being superior in terms of remission from binge eating/ purging at follow up (N = 1; n = 67; RR = 0.70; 95% CI, 0.50 to 0.97; NNT = 4; 95% CI, 3 to 25). I
Effect of treatment on the frequency of binge eating and purging There is evidence from five trials that the combination of CBT and antidepressants is equivalent to or better than CBT alone with respect to the reduction of binge eating or purging by the end of treatment:
• There is limited evidence suggesting that there is a clinically significant difference between CBT and a combination of CBT and an antidepressant with combination being superior on mean frequency of
16 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
154 CONFIDENTIAL Second draft for consultation September 2003
binge eating by the end of treatment (N = 5; n = 185; SMD = 0.38; 95% CI, 0.09 to 0.68). I
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT and a combination of CBT and an antidepressant on mean frequency of purging by the end of treatment (N = 5; n = 157; SMD = 0.16; 95% CI, -0.16 to 0.48). I
There is insufficient evidence to determine whether CBT differs from the combination of CBT and antidepressants with respect to the frequency of binge eating or purging at post-treatment follow-up. I
There is insufficient or no evidence to determine whether PSH or focal supportive psychotherapy differ from the combination of these psychological treatments and antidepressants with respect to the frequency of binge eating or purging by either the end of treatment or post-treatment follow-up. I
There is insufficient evidence to determine whether dietary counselling differs from the combination of dietary counselling and antidepressants with respect to the frequency of binge eating or purging by either the end of treatment or post-treatment follow-up. I
Other effects of treatment There is insufficient or no evidence to determine whether CBT differs from the combination of CBT and antidepressants on depression or general psychiatric symptoms by either the end of treatment or post-treatment follow- up. I
There is insufficient or no evidence to determine whether PSH, focal supportive psychotherapy or dietary counselling differ from the combination of these psychological treatments and antidepressants with respect to depression or general psychiatric symptoms by either the end of treatment or post-treatment follow-up. I
Acceptability of treatment There is insufficient evidence to determine whether CBT is more, or less, acceptable to people with bulimia nervosa when compared with the combination of CBT and antidepressants. I
There is insufficient evidence to determine whether PSH or focal supportive psychotherapy are more, or less, acceptable to people with bulimia nervosa when compared with the combination of these psychological treatments and antidepressants. I
There is insufficient evidence to determine whether dietary counselling is more, or less, acceptable to people with bulimia nervosa when compared with the combination of these dietary counselling and antidepressants. I
155 CONFIDENTIAL Second draft for consultation September 2003
7.4.5 Antidepressant drugs versus the combination of antidepressant drugs and psychological treatment
7.4.5.1 Treatments reviewed The following antidepressant drugs were included: • SSRIs (fluoxetine)
• Tricyclic antidepressants (desipramine, imipramine).
The following psychological treatments were included: • CBT
• Focal supportive psychotherapy
• Dietary counselling
• PSH.
7.4.5.2 Studies considered The review team used the existing Cochrane review ‘Antidepressants versus psychological treatments and their combination for bulimia nervosa’ (Bacaltchuk, Hay & Trefiglio, 2003) as the starting point for the present analysis.
All five trials comparing an antidepressant to a combination of an antidepressant and a psychological treatment, used in the existing Cochrane review, were included in this section (AGRAS1992, GOLDBLOOM1997, LEITENBERG1994, MITCHELL1990, WALSH1997). In addition, three trials (MITCHELL2001A, JACOBI2002, WALSHSUBMIT) found during the search for new evidence, were included here. Thus, 8 RCTs involving 694 participants were included.
Of the included trials, six used CBT as the psychological treatment (AGRAS1992, GOLDBLOOM1997, JACOBI2002, LEITENBERG1994, MITCHELL1990, WALSH1997) and two used self-help (MITCHELL2001A, WALSHSUBMIT). In addition, WALSH1997 used a focal supportive psychotherapy group.
Full details of the studies included in the guideline and the reasons for excluding studies are given in appendix 17.
7.4.5.3 Evidence statements17
17 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta-
156 CONFIDENTIAL Second draft for consultation September 2003
The data were analysed by comparing any antidepressant to any antidepressant combined with a specific psychological treatment.
Effect of treatment on remission from binge eating/ purging There is insufficient evidence to determine whether antidepressants differ from the combination of CBT and antidepressants with respect to remission from binge eating or purging by either the end of treatment or post-treatment follow-up. I
There is insufficient evidence to determine whether antidepressants differ from the combination of PSH or focal supportive psychotherapy and antidepressants with respect to remission from binge eating or purging by either the end of treatment or post-treatment follow-up. I
Effect of treatment on the frequency of binge eating and purging There is evidence from five trials favouring the combination of CBT and antidepressants over antidepressants alone in terms of the frequency of binge eating and purging by the end of treatment:
• There is strong evidence suggesting that there is a clinically significant difference between antidepressants and a combination of CBT and an antidepressant with the combination being superior on mean frequency of binge eating by the end of treatment (N = 4; n = 133; SMD = 0.55; 95% CI, 0.21 to 0.90). I
• There is limited evidence suggesting that there is a clinically significant difference between antidepressants and a combination of CBT and an antidepressant with the combination being superior on mean frequency of purging by the end of treatment (N = 5; n = 141; SMD = 0.49; 95% CI, 0.15 to 0.83). I
There is insufficient evidence to determine whether antidepressants differ from the combination of CBT and antidepressants with respect to the frequency of binge eating or purging at post-treatment follow-up. I
There is insufficient or no evidence to determine whether antidepressants differ from the combination of PSH or focal supportive psychotherapy and antidepressants with respect to the frequency of binge eating or purging by either the end of treatment or post-treatment follow-up. I
Other effects of treatment There is insufficient or no evidence to determine whether antidepressants differ from the combination of CBT and antidepressants on depression or
analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
157 CONFIDENTIAL Second draft for consultation September 2003 general psychiatric symptoms by either the end of treatment or post- treatment follow-up. I
There is insufficient or no evidence to determine whether antidepressants differ from the combination of PSH or focal supportive psychotherapy and antidepressants on depression or general psychiatric symptoms by either the end of treatment or post-treatment follow-up. I
Acceptability of treatment There is insufficient evidence to determine whether antidepressants are more, or less, acceptable to people with bulimia nervosa when compared with the combination of CBT and antidepressants. I
There is insufficient or no evidence to determine whether antidepressants are more, or less, acceptable to people with bulimia nervosa when compared with the combination of PSH or focal supportive psychotherapy and antidepressants. I
Tolerability of treatment There is insufficient evidence to determine whether antidepressants have more, or less, side effects in people with bulimia nervosa when compared with the combination of CBT and antidepressants. I
There is insufficient or no evidence to determine whether antidepressants have more, or less, side effects in people with bulimia nervosa when compared with the combination of PSH or focal supportive psychotherapy and antidepressants. I
7.4.6 Clinical summary Four points emerge from these analyses and the studies upon which they are based. First, there have been few comparisons of psychological and pharmacological treatments, and their combination, with the result that any practice recommendations must be tentative. Second, few studies have included post-treatment follow-up periods, a problem with almost all the studies that have used drugs. Third, the comparisons of CBT-BN with antidepressant drugs indicate that CBT-BN is the more potent treatment. Fourth, the combination of CBT with antidepressant drugs is superior to antidepressant drugs on their own. Lastly, there is limited evidence that suggests that adding antidepressant drugs to CBT may enhance the effect of CBT on binge eating.
7.4.7 Clinical practice recommendations
7.4.7.1 For people with bulimia nervosa who are not responding to cognitive behaviour therapy - bulimia nervosa, the addition of an antidepressant drug should be considered as this may improve outcome. (B)
158 CONFIDENTIAL Second draft for consultation September 2003
7.5 Management of the physical aspects of bulimia nervosa
7.5.1 Introduction For the majority of people with bulimia nervosa, serious medical complications are rare. Severe problems can arise when there is co-morbidity with diabetes. In addition, some of the compensatory behaviours can be toxic (herbal preparations, high dose analgesics, diuretics and Ipecac). The longer- term physical risks include damage to teeth from vomiting and overuse of carbonated drinks and damage to the gut from laxative abuse.
The crude mortality rate (CMR) of bulimia nervosa was 0.4% (11 deaths in 2692 patients), when two meta-analyses were combined (Keel & Mitchell, 1997; Nielsen, 2001). No information is available as to the distribution of causes of death.
In one meta-analysis (Nielsen, 2001), the overall aggregate SMR of bulimia nervosa in studies with 5-11 years of follow-up was 7.4 (95% CI, 2.9 to 14.9).
In addition to the general nutritional deficiency that may be present in bulimia nervosa, there may be specific nutritional problems related to meal content and patterning and purging.
The physical management of bulimia nervosa therefore falls (with the exceptions described below) into normal medical practice with the provisos set out in the introduction and primary care section concerned with detection and appropriate investigations. Good history taking and skilled empathic interviewing are necessary to elicit the full range of compensatory behaviours as these are often secretive, shameful and can be illegal.
The physical risk assessment screen used for anorexia nervosa may also be considered for bulimia nervosa.
7.5.2 Physical complications of binge eating/ purging The physical consequences of bulimia nervosa result largely from the compensatory behaviours of vomiting, laxative and diuretic abuse. For the purposes of this section the following complications were considered: fluid and electrolyte disturbances, damage to the gastrointestinal tract and dental complications. Evidence for the management of these difficulties is limited to descriptive reviews, other guidelines and knowledge of common medical practice.
Fluid and electrolyte disturbances Fluid and electrolyte disturbances occur in bulimia nervosa and relate to the severity of symptoms and the general nutritional status. Common abnormalities include:
159 CONFIDENTIAL Second draft for consultation September 2003
• dehydration
• hypokalaemia
• hypochloraemia
• alkalosis.
Dehydration can cause volume depletion and consequently low blood pressure and a rapid pulse. Patients can complain of dizziness because of orthostatic hypotension and weakness. In extreme cases renal function can be compromised. Secondary hypoaldosteronism can lead to rebound, fluid retention and peripheral oedema when laxatives and diuretics are withdrawn (Mitchell, 1988).
Low potassium causes weakness in all muscle and most worryingly cardiac arrhythmias. Renal function can also be affected. Metabolic alkalosis may augment potassium depletion. Diuretic abuse, particularly thiazide and loop diuretics, can produce marked potassium and sodium depletion.
Low sodium and magnesium levels occur less commonly but both have potentially serious consequences. If severe the former may cause central nervous system disturbances, whilst the latter results in muscle weakness, cardiac arrhythmias and mood changes. Low magnesium is also associated with other abnormalities such as hypocalcaemia and hypokalaemia. Clinicians should consider the presence of low magnesium levels in the face of refractory hypokalaemia.
In general, these abnormalities settle with cessation of purging behaviours. If needed, oral rather than IV supplementation (with substances such as rehydrate in severe cases) is advised. Advice from a physician and/or paediatrician may be necessary if there is severe metabolic disturbance (Connan, Lightman & Treasure, 2000). Very rarely patients may require hospital admission to manage severe purging behaviour.
Gastro–intestinal damage Loss of the gag reflex and gastro-oesophageal reflux has been found in patients with bulimia nervosa. Frequent and severe vomiting can cause more serious complications of oesophageal tears, perforation, and oesophagitis. These may require surgical intervention.
Laxative abuse The prevalence of daily laxative abuse in patients with bulimia nervosa has been reported as up to 20% (Mitchell et al., 1985) and 75% of patient report using laxatives at some point (Fairburn & Cooper, 1984). Patients usually take laxatives with the mistaken assumption that there is a consequent reduction in calorie absorption. The fluid loss that results from laxative action on the
160 CONFIDENTIAL Second draft for consultation September 2003
colon can provide a feeling of emptiness and weight loss. However laxatives have been shown to be an ineffective method of reducing calorie absorption (Bo-Linn et al., 1983). The majority of patients use stimulant rather than bulk laxatives. The former have been shown to cause degeneration of the colonic nerve supply (Oster, 1980).
Laxative abuse carries the acute complication of electrolyte and fluid disturbances and can be particularly dangerous in low weight individuals. Abrupt cessation of laxatives in those who are taking them regularly can result in reflex fluid and sodium retention, and consequent weight gain, and oedema. This can increase patient anxiety and reluctance to curtail the use of laxatives. To avoid this effect a gradual reduction in laxative use is advised.
Long term excessive laxative use can decrease the motility of the colon. This results in constipation and more worryingly an atonic cathartic colon. Recommended treatment of the constipation is regular food intake, bulk laxatives, adequate fluids and exercise (Robinson, 2000). Some patients may find these measures uncomfortable. Rarely surgical intervention may be required.
Dental complications Dental erosion is the most common oral problem in patients with eating disorders that engage in self-induced vomiting. Erosion of the enamel of the tooth surface is caused by exposure to gastric acid. The pattern of erosion is distinctive and may indicate the aetiology (i.e., vomiting) to dental practitioners. Teeth may become discoloured and change shape. Tooth sensitivity (pain on eating hot or cold meals) has also been reported.
Detailed Clinical Guidelines on the Diagnosis and Prevention of Dental Erosion are available (http://www.rcseng.ac.uk/dental/fds/clinical_guidelines). These guidelines stress that recognition and management of the underlying aetiology (i.e., the eating disorder) is essential. It is noted that referral by dental practitioners to the general practitioner for treatment may be required and that “initiation of medical help is a sensitive undertaking”. Dental practitioners are advised to monitor the erosion, to offer advice and provide treatment to prevent further erosion and improve appearance. Restorative treatment may prove difficult in the face of ongoing vomiting.
For clinicians involved with the management of the eating disorder, it is important to advise patients on the dental consequences and refer to a dental practitioner. Although any treatment of the eating disorder will aim to limit vomiting, patients should be given the following advice:
• Brushing teeth after vomiting should be avoided as it may increase tooth damage
161 CONFIDENTIAL Second draft for consultation September 2003
• Mouth rinsing after vomiting with water and sodium bicarbonate (or other non-acid mouth wash) will neutralise the acid environment
• Visit the dentist regularly
• Fluoride mouth rinses and toothpastes may be helpful for desensitisation.
A high intake of acidic foods may increase dental erosion. These foods include fruit, fruit juice, carbonated drinks, pickled products, yoghurt and some alcoholic drinks. Dental practitioners guidelines recommend detailed dietary advice, which these patients may find difficult to follow. This advice includes: • Limiting acidic foods and drinks to mealtimes
• Finishing meals with alkaline foods (e.g., milk or cheese)
• Avoiding acid foods and drinks last thing at night
• Avoiding habits such as prolonged sipping, holding acidic beverages in the mouth and “frothing” prior to swallowing
• Avoiding toothbrushing after acidic substances
• Chewing sugar-free gum after meals to stimulate salivary flow (although this may cause increased gastric secretions).
Whilst this guidance will contribute to a reduction in dental erosion, it may not always be possible to follow it, particularly in the early stages of the treatment plan. For example, the consumption of some acidic foods (yoghurt and fruit) may be a necessary stage in stabilising a chaotic eating pattern, while encouraging other strategies to limit acid in the mouth.
Other physical consequences Fasting and binge eating foods high in refined carbohydrates, especially if this is followed by vomiting can lead to high levels of insulin release with large fluctuations in blood sugar levels (Johnson, Jarrell, Chupurdia & Williamson, 1994). This may disrupt the appetite control mechanisms and the utilisation and deposition of energy
Serotonin is implicated in appetite regulation (there may be a particular role in carbohydrate balance) (Leibowitz & Alexander, 1998)), disruptions in serotonin levels may be affected by the impact of insulin on its precursor, tryptophan, the levels of which can be adversely affected by dieting, particularly in women. In turn acute tryptophan depletion may lead to an increase in calorie intake and irritability in bulimia nervosa (Weltzin,
162 CONFIDENTIAL Second draft for consultation September 2003
Fernstrom, Fernstrom, Neuberger & Kaye, 1995) and may be related to decreased mood, increased rating in body image concern and subjective loss of control of eating in people who have recovered from bulimia nervosa (Smith, Fairburn & Cowen, 1999). People with bulimia nervosa appear to have dysregulation in the release of factors involved in the peripheral response to food such as cholecystokinin (Geracioti & Liddle, 1988) and ghrelin. (Tanaka et al., 2002). There is some suggestion of dysregulation in the systems that are involved in longer-term energy balance such as leptin (Brewerton, Lesem, Kennedy & Garvey, 2000; Monteleone et al., 2000; Jimerson, 2002), which may persist after recovery (Jimerson, 2002).
In summary, people with bulimia nervosa may have abnormal levels of tryptophan and central serotonin along with a potential disruption in both the acute and the longer-term mechanisms controlling hunger and satiety.
7.5.3 Clinical practice recommendations
7.5.3.1 Patients with bulimia nervosa who are vomiting frequently or taking large quantities of laxatives (and especially if they are also underweight) should have their fluid and electrolyte balance assessed. (C)
7.5.3.2 When electrolyte disturbance is detected, it is usually sufficient to focus on eliminating the behaviour responsible. In the small proportion of cases where supplementation is required to restore the patient’s electrolyte balance, oral rather than intravenous administration is recommended, unless there are problems with gastro intestinal absorption. (C)
7.5.3.3 Where laxative abuse is present, patients should be advised to gradually reduce laxative use and informed that laxative use does not significantly reduce calorie absorption. (C)
7.5.3.4 Patients with bulimia nervosa should have regular dental reviews. (C)
7.5.3.5 Patients who are vomiting should be given appropriate advice on dental hygiene including avoiding brushing after vomiting, rinsing with a non-acid mouth wash after vomiting, and measures to reduce an acid oral environment (limiting acidic foods). (C)
7.5.4 Somatic interventions One of the risk domains for bulimia nervosa relates to dieting and weight control. People with bulimia nervosa frequently have a personal or family history of obesity. One of the important maintaining factors is the perceived need (or wish) to control weight. There has been recent interest in the use of somatic interventions including exercise and massage. In the case of some
163 CONFIDENTIAL Second draft for consultation September 2003
exercise, this relates to the evidence that it may be of benefit in improving mood (Babyak et al., 2000). It is also frequently used in the treatment of obesity. However, current practice is not underpinned by a strong evidence base. For this section two interventions were considered: • Exercise
• Massage.
7.5.4.1 Studies considered The review team conducted a new systematic search for RCTs that assessed the efficacy of exercise or massage for people with bulimia nervosa.
Two trials met the eligibility criteria set out by the GDG. One compared exercise with a non bulimia nervosa focused form of CBT (SUNDGOTBORGEN2002), and one trial compared massage in addition to ‘standard care’ with ‘standard care’ alone (FIELD1998A). Thus, two trials involving 88 participants were included in this section.
Full details of the studies included in the guideline and the reasons for excluding studies are given in appendix 17.
7.5.4.2 Evidence statements18
Exercise
Effect of treatment on the frequency of binge eating and purging There is limited evidence suggesting that there is a clinically significant difference between exercise and a non bulimia nervosa focused form of CBT with exercise being superior in terms of mean frequency of binge eating by 18 months post-treatment follow-up (N = 1; n = 26; SMD = -0.83; 95% CI, -1.64 to -0.02). I
There is insufficient evidence to determine whether exercise differs from CBT with respect to the frequency of purging by follow-up. I
Acceptability of treatment There is insufficient evidence to determine whether exercise is more, or less, acceptable to people with bulimia nervosa than CBT. I
Massage
18 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
164 CONFIDENTIAL Second draft for consultation September 2003
Other effects of treatment There is limited evidence to suggest that the addition of massage to 'standard care' is superior to ‘standard care’ alone in terms of mean scores on the Eating Disorders Inventory by the end of treatment (n = 24; FIELD1998a). I
Acceptability of treatment There is insufficient evidence to determine whether massage is more, or less, acceptable to people with bulimia nervosa than ‘standard care’. I
7.5.5 Clinical summary The use of somatic treatments (exercise and massage) is not frequent in the UK. There is limited evidence from one small RCT suggesting that exercise may reduce the frequency of binge eating when compared to a non bulimia nervosa focused form of CBT. In addition, massage may produce some benefit over and above ‘standard care’. However, there is currently no RCTs comparing exercise or massage with psychological interventions shown to be effective for the treatment of people with bulimia nervosa. Until a more substantial evidence base is established it is unlikely that either will form part of the routine practice of eating disorder services in the UK.
7.5.6 Concurrent physical conditions A number of physical conditions present particular problems when they present co-morbidly with bulimia nervosa. The problems presented by these conditions are set out below, followed by a summary on approaches to the management of the problems as they present.
Diabetes Diabetes is over-represented in people with bulimia nervosa. There is a three- fold increase in the odds ratio for bulimia nervosa (OR = 2.9; 95% CI, 1.0 to 8.4) and a significant two-fold increase in odds ratio for both EDNOS and sub- threshold eating disorders in people with Type 1 Diabetes (Nielsen, 2002). It has been suggested that Type 1 Diabetes may precede the eating disorder (Neilson et al., 2002). Omission or intentional under-dosing of insulin so called ‘insulin-purging’ is increased (OR = 12.6; 95% CI, 7.8 to 21.1). This leads to poor control of blood sugar and an increased risk of physical complications. Increased retinopathy is also found where eating disorders co- exist with Type 1 Diabetes (OR = 4.8; 95% CI, 3.0 to 7.8).
In the largest (n = 91) longitudinal study (4 years) of a cohort of people with eating disorders and Type 1 Diabetes, disturbed eating persisted in 60% of cases and was associated with a 3-fold increase in retinopathy (Rydall, Rodin, Olmsted, Devenyi & Daneman, 1997). In a cohort of young diabetics (n = 43 males, n = 33 females) followed over 10 years; those with behavioural problems in adolescence had a worse outcome (Bryden, Peveler, Stein, Neil, Mayou & Dunger, 2001).
165 CONFIDENTIAL Second draft for consultation September 2003
The presence of Type 1 Diabetes presents challenges not only for physical management but also for psychological treatment. One of the goals of CBT for bulimia nervosa is to relax control over eating and this can conflict with the nutritional advice given to diabetics (Peveler, Davies, Mayou, Fairburn & Mann, 1993). On the other hand it is most important that these patients are helped to overcome their eating disorder given the associated physical complications. There are guidelines for adapting CBT-BN for these patients (Peveler & Fairburn, 1992).
One empirical study, an RCT, compared psycho-education with ‘standard care’ for people with bulimia nervosa and Type 1 Diabetes (Olmsted, 2002). In this study, 212 women attending a paediatric diabetes clinic were screened for signs of eating disturbance. One hundred and thirty passed the screening and were invited to participate in the intervention phase of the study. Eighty-five participants were randomised to psychoeducation or ‘standard care’. Assessments were conducted before and after treatment and at 6-month follow-up.
An intention-to-treat group by time multivariate analyses of variance (MANOVA) indicated significant reductions following psychoeducation on the Restraint and Eating Concern subscales of the Eating Disorder Examination (EDE) and on the Drive for Thinness and Body Dissatisfaction subscales of the Eating Disorder Inventory (EDI), but no improvement in frequency of purging by insulin omission (mean 1.4 insulin omission days at baseline and 1.3 at 6 month follow up) or haemoglobin A1c levels (mean at baseline 9.2 and at 6 months follow up 9.3). Psychoeducation was associated with reductions in eating disturbance, but not with improved metabolic control.
7.5.7 Clinical summary The conclusions to be drawn from the review are that females with Type 1 Diabetes have an increased risk of bulimia nervosa and EDNOS and that poor adherence to insulin treatment is frequent in eating disordered Type 1 Diabetes individuals. This comorbidity complicates psychological interventions. Psychoeducation may have some limited benefit on eating disorder symptoms but not on diabetic control. In the management of people with bulimia nervosa and Type 1 Diabetes, close liaison and a shared knowledge base between the eating disorder and diabetes teams is essential.
7.5.8 Clinical practice recommendations
7.5.8.1 Treatment of both subthreshold and clinical cases of bulimia nervosa in people with diabetes is essential because of the greatly increased physical risk in this group. (C)
166 CONFIDENTIAL Second draft for consultation September 2003
7.5.8.2 Patients with Type 1 Diabetes and bulimia nervosa should have intensive regular physical monitoring as they are at high risk of retinopathy and other complications. (C)
Pregnancy and the postnatal period The peak age of prevalence of eating disorders is in the childbearing years. Women with bulimia nervosa are at risk of unplanned pregnancies due to mistaken beliefs about fertility in the presence of oligomenorrhoea (Morgan, Lacey & Sedgwick, 1999). and also because many women take an oral contraceptive pill which they inadvertently vomit. In the majority of cases bulimic symptoms will improve during pregnancy and for a period of time after the birth. Lemberg and Phillips (1989) found that the fear of losing control over weight gain was one of the most common worries during pregnancy. Most women had not mentioned their eating disorder to their obstetricians.
In a small case series a higher than expected number of complications were were reported (Lacey & Smith, 1987). It is not certain whether such complications can be ascribed to the core symptoms/behaviours of bulimia nervosa or whether they are caused by co-morbidity such as alcohol and drug abuse found among these patients (Key, Mason & Bolton, 2001; Abraham, 1998). Women with bulimia nervosa are at risk of delivering low birth weight infants, and have higher rates of Caesarean sections and perinatal problems (Franko et al., 2001). A two-fold increase in miscarriage rate was reported for a bulimic population (Abraham, 1998; Willis & Rand, 1988). In Waugh and Bulik’s study (1999) the children of mothers with eating disorders had a significantly lower birth weight and length.
Eating disorder symptomatology has been found to increase in the postpartum in the majority of studies on small clinical samples of women with eating disorders (Lacey & Smith, 1987; Lemberg & Phillips, 1989; Morgan, Lacey & Sedgwick, 1999; Stein & Fairburn, 1996). Mothers had difficulties in maintaining breast-feeding and in making positive comments about food and eating at mealtimes, they were less likely to cook or eat with their children (Waugh & Bulik, 1999). The infants of mothers with eating disorders have a lower mean weight than controls. The infants’ weight was associated with conflict during mealtimes and mothers’ concerns about their own weight (Stein, Woolley, Cooper & Fairburn, 1994). Moreover, in a small prospective study on women with eating disorders, mothers used a less regular feeding schedule for their infants and used food more for non- nutritive purposes (Agras, Hammer & McNicholas, 1999). They also had higher concerns about their daughters’ weight from the age of 2 onwards. Lacey & Smith (1987) reported that 15% of mothers in their series reported attempting to "slim down" their babies.
167 CONFIDENTIAL Second draft for consultation September 2003
Also mothers report increase problematic interactions, for example, increased irritability and inability to cope with the child’s demands while binge eating. Children of mothers with eating disorders showed more negative affect, mainly sadness and crying (Agras et al., 1999). In addition to emotional problems, language and speech problems have been reported shown (Franzen & Gerlinghoff, 1997). Also there have been reports of inappropriate involvement of children in their parent’s illness, with “parenting” on the part of the child and role-reversal (Brandes & Lackstrom, 1993; Woodside, Shekter- Wolfson, Garfinkel & Olmsted, 1995).
7.5.9 Clinical summary Although for many women with bulimia nervosa pregnancy can pass uneventfully, for some pregnancy and the post partum period can present considerable challenges. For this at risk group four points should be considered. First, women with bulimia nervosa may have more complications during pregnancy and the children who are smaller at birth. Secondly, they may also have problematic interactions feeding their infants and children and use food for instrumental rather nutritive reasons. Thirdly, they tend to be more concerned about their child’s weight and shape. Finally they may have more problematic interactions with their children.
7.5.10 Clinical practice recommendations
7.5.10.1 Pregnant women with bulimia nervosa require careful monitoring throughout the pregnancy and in the post-partum period. (C)
7.6 Service level interventions
7.6.1 Introduction In the UK, most people presenting with bulimia nervosa do so in primary care and over the last twenty years the numbers doing so have increased markedly (Turnbull, Ward, Treasure, Jick & Derby, 1996). Some patients may receive adequate help there. For many people interventions may consist of the prescription of antidepressant drugs or the reading of books or using similar self-help materials with or without professional guidance. The number referred onto secondary care services, whether general mental health services or specialist eating disorder services is not known. However, the filters on the pathway to more specialised secondary and tertiary care are complex. As indicated in the introduction, the availability of appropriate expertise is variable and may be the greatest single determinant of the kind of help an individual with bulimia nervosa receives. Whether an individual receives appropriate and effective services depends upon many factors including the existence and availability of such services. Within secondary services, bulimia nervosa is almost always treated on an outpatient basis. Inpatient treatment is usually reserved for the management of severe psychiatric co-morbidity and physical complications. Special inpatient programmes for the treatment of
168 CONFIDENTIAL Second draft for consultation September 2003 bulimia nervosa are unusual in the UK, although they are more common in the rest of Europe.
7.6.2 Effective service configurations The treatment of bulimia nervosa may involve a range of interventions from self-help in primary care through to, in occasional cases, specialist inpatient treatment. Treatments vary in type and in the service level – outpatient, day patient or inpatient – at which they are delivered. The idea of “stepped care” has been widely advocated (Fairburn & Peveler, 1990; Dalle Grave, Ricca & Todesco, 2001). This idea proposes that patients should be offered simpler and less expensive interventions first and that more complex and expensive interventions should be reserved for those who have not benefited. The related ideas of sequencing and integration of different types of treatment has also been advocated (Garner & Needleman, 1997).
Unfortunately there have been no systematic comparisons of outcome in different service levels. The few comparisons are not randomised and are difficult to interpret. (Williamson, Prather, Bennett, Davis, Watkins & Grenier, 1989). The great majority of reports are of outpatient treatments. All of the current evidence based therapies for bulimia nervosa are designed to be delivered in an outpatient setting.
The place of inpatient treatment for bulimia nervosa is not supported by research evidence. Special inpatient and day patient treatment regimes have been described (Zipfel et al., 2002). However, in these cases the role of admission would seem to be advocated mainly in relation to extreme severity, co-morbidity or suicidal risk. There are some reports on special treatment programmes for severe bulimia nervosa complicated by self-harm, substance abuse and similar behaviours in patients who often also fulfil criteria for borderline personality disorder (Lacey, 1997). Such programmes have not been adequately evaluated and most such patients can be managed on an outpatient basis.
7.6.3 Satisfaction with services The area of patient user and carer perceptions has been the focus of some research in the area of eating disorders. Much of this work has been focused on perceptions of inpatient treatment for anorexia nervosa or on mixed or poorly defined populations of people with eating disorders, therefore it should be treated with caution in the specific context of bulimia nervosa. The reader is referred to the chapter on anorexia nervosa for a fuller review of the evidence base for satisfaction with services.
7.6.4 Death, suicide and self-harm, and the relationship to service setting Bulimia nervosa is associated with a significant risk of deliberate self-harm behaviour (Anderson, McIntosh, Joyce & Bulik, 2002; Favaro & Santonastaso, 1997). The results of studies of overall mortality suggest a possible increase in bulimia
169 CONFIDENTIAL Second draft for consultation September 2003
nervosa although the extent is not clear (Nielsen, 2001). However, it is certainly the case that patients may die prematurely as a result of the disorder or its complications. People with bulimia nervosa may be offered treatment in outpatient, day patient or inpatient settings with the aim of ameliorating the risk of death through complication or suicide, or to contain self-harm.
7.6.5 Clinical practice recommendations
7.6.5.1 The great majority of patients with bulimia nervosa should be treated in an outpatient setting. (C)
7.6.5.2 For patients with bulimia nervosa who are at risk of suicide or severe self-harm, admission as an inpatient or a day patient or the provision of more intensive outpatient care, should be considered. (C)
7.6.5.3 Psychiatric admission for people with bulimia nervosa should normally be undertaken in a setting with experience of managing this disorder. (C)
7.7 Predicting the outcome of treatment and recovery from bulimia nervosa
Predicting response to treatment in bulimia nervosa is difficult and currently poorly understood. A particular complication is that co-morbidity in bulimia nervosa is common and it is often unclear whether patients presenting with such problems should have a standard treatment or how such treatment should be modified, enhanced or complemented. Even with the best treatments currently available, up to 50% of all bulimia nervosa patients may not respond adequately. In principle, it is desirable not to allocate patients to initial interventions that are unlikely to be of benefit as this may demoralise patients and be wasteful of resources. It is therefore important to identify which factors may influence outcome.
Two papers provide narrative reviews of predictors of outcome for bulimia nervosa. Keel and Mitchell (1997) reviewed data on 60 studies, which assessed outcome at least 6 months after presentation. They concluded that there were few consistent factors although ‘personality traits such as impulsivity may contribute to poor outcome’. Vaz (1998) reviewed data on ‘individual, environmental and therapeutic’ factors. In the reviews so far conducted, a distinction has not always been made between studies employing a mixed eating disorder population and those of a single diagnosis. A number of papers review various concurrent psychopathology and its relationship to outcome in eating disorder populations (Holderness et al., 1994; Herzog et al., 1996; Rosenvinge et al., 2000). Bell (2002) reviewed the evidence specifically for bulimia nervosa and concluded that there is no consistent relationship between psychiatric comorbidity and outcome for bulimia nervosa, but the
170 CONFIDENTIAL Second draft for consultation September 2003
presence of co-existing impulse control problems or cluster B personality disorder is probably associated with poorer outcome.
A wide range of potential predictors of outcome have been studied, often those which are routinely or most easily collected at pre-treatment. Some predictors, such as readiness for change, which may be important, are rarely measured. The impact of a potential predictor will vary at different points in the course of a disorder and few studies address this (Garner et al., 1990; Fahy & Russell, 1993; and those of the Oxford group are exceptions). A number of different factors may influence speed of response to treatment, outcome at end of treatment, outcome at follow-up, relapse and chronicity. Most studies assess relationships between pre-treatment measures and outcome at end of treatment; some report on outcome at follow-up. Factors which predict end- of-treatment outcome may not predict outcome at follow-up.
For the purposes of the guideline, a literature review was carried out using PsycLIT and MEDLINE, which augmented the search of clinical trials already performed as part of the review of the effectiveness of treatments. A wide range of studies (cohort and treatment studies) of varying quality were considered. Studies of mixed eating disorder populations were excluded unless separate data for anorexia nervosa and bulimia nervosa were reported Sample size (n ≥ 50) was selected as the key inclusion criterion as it was not possible to derive robust measures of study quality, and both positive and negative findings were included. All variables reported in the studies were included in the analysis. One exception was made to the key inclusion criteria; studies with positive findings and a sample size of less than 50 were included if the follow-up was 1 year or over. A total of 60 studies contributed to the final analysis with sample sizes ranging from 17-647. All but two studies (Fairburn et al., 2003; Stice & Agras, 1998) were of outcome in response to specific treatments.
The wide variability in method did not permit a meta-analysis. A detailed analysis by intervention was also beyond the scope of this review, though the data suggest a difference between interventions with respect to the frequency of pre-treatment binge eating and purging. A detailed analysis of the number of participants leaving a study early, as an outcome measure, or an analysis by follow-up was also beyond the scope of the review.
Outcome measures varied considerably across studies (abstinence, diagnostic status, changes or percentage reduction in the frequency of binge eating/ purging). Some studies also examined co-morbid disorders as measures of outcome. For consistency all findings are expressed in relation to poor outcome. A wide range of statistical techniques was used in the studies under review but an assessment of their quality and appropriateness was not undertaken.
171 CONFIDENTIAL Second draft for consultation September 2003
Based on the number of studies showing a positive result relative to the number showing a negative result for each variable examined, a number of possible predictors emerged (for full details see Appendix 11). There is less robust evidence for predictors of outcome in bulimia nervosa than anorexia nervosa but those with lower motivation for change prior to treatment, had poorer outcomes. This was also the case for concurrent substance misuse and a history of obesity. Higher levels of binge eating and purging prior to treatment were associated with poorer outcome in abbreviated treatments. Continuing bulimic behaviours at the end of treatment is also associated with poorer long-term outcome. It is possible that treatment that is terminated before abstinence is achieved, may compromise long-term recovery but in the absence of firm evidence this remains speculative. There is some evidence to suggest that patients with concurrent borderline features do less well but this needs further research.
Those who respond poorly to treatment have a wide range of problems at the end of treatment including poor social adjustment, lack of a stable relationship, and depressive features. Disordered eating attitudes, body dissatisfaction and drive for thinness at the end of treatment and their associated behaviours (i.e., dietary restriction and vomiting) were associated with poorer outcome in the longer term. Outcome for these patients may be improved by addressing these issues within treatment.
7.7.1 Clinical practice recommendation
7.7.1.1 Healthcare professionals should be aware that patients with bulimia nervosa who have poor impulse control including substance misuse are unlikely to respond to a standard programme of treatment. As a consequence treatment should be adapted to the problems presented. (C)
172 CONFIDENTIAL Second draft for consultation September 2003
8 Treatment and management of atypical eating disorders (eating disorders not otherwise specified) including binge eating disorder
8.1 Introduction In the absence of research evidence to guide the management of atypical eating disorders (Fairburn & Harrison, 2003) (also termed Eating Disorders Not Otherwise Specified, EDNOS; APA, 1994), other than binge eating disorder (BED), it is recommended that clinicians treat these patients following the principles advocated for treating the eating disorder that their eating problem most closely resembles.
In addition to the evidence base for anorexia nervosa and bulimia nervosa covered in other chapters of this guideline, BED has a growing evidence base relating to CBT, IPT and self-help approaches. There is also emerging research on antidepressants and appetite suppressants. To date, the research has focused primarily on the treatment of cases with comorbid obesity. It is not known whether equivalent findings would be obtained with patients who do not have obesity. Similar treatment approaches may well be helpful in adolescent cases though this has not been researched. There is no evidence about service setting for the treatment of BED. 8.2 Psychological interventions
8.2.1 Introduction Psychological treatments for atypical eating disorders have been little studied, although patients with these disorders comprise a very significant proportion of those who present for treatment in the NHS (Fairburn & Harrison, 2003). The past 10 years has seen a growing interest in the treatment of BED with the development of specialised forms of CBT and IPT along with simplified forms of DBT. Behavioural weight management programmes have also been evaluated as a form of treatment for BED when it co-occurs with obesity.
8.2.2 Current practice As indicated above, although patients with atypical eating disorders comprise a significant percentage of patients with eating disorders in specialist service settings, little is known about their current management. It is clearly a priority that treatment of these patients receives research attention.
173 CONFIDENTIAL Second draft for consultation September 2003
8.2.3 Psychological treatments compared with wait-list controls
8.2.3.1 Treatments reviewed The treatments included in this section are for BED only, usually in the presence of obesity.
The following treatments were included:
• Behavioural weight control (BWC)
• Cognitive behaviour therapy for binge eating disorder (CBT-BED)
• Interpersonal psychotherapy for binge eating disorder (IPT-BED)
• Simplified dialectical behaviour therapy (simplified DBT).
The Psychological Topic Group established definitions for each treatment (see Glossary). Two members of the Topic Group assessed each study for eligibility and classified each psychological treatment. Where disagreements arose, they were resolved by discussion.
8.2.3.2 Studies considered19 The review team conducted a new review of psychological treatments compared to wait-list control for BED. Seven trials met the eligibility criteria set by the Psychological Topic Group (AGRAS1995, ELDREDGE1997, GORIN2001, REEVES2001, TELCH1990, TELCH2001, WILFLEY1993). Of these, four used CBT-BED as the active treatment (AGRAS1995, ELDREDGE1997, GORIN2001, TELCH1990, WILFLEY1993), one used BWC (REEVES2001), one used simplified DBT (TELCH2001), and one used IPT- BED (WILFLEY1993). Thus, seven RCTs comparing a psychological treatment with a wait-list control group, involving 432 participants, were included in this section.
Full details of studies included in this review and reasons for excluding studies are given in Appendix 17.
8.2.3.3 Evidence statements20 The level of evidence (I, IIb, IIb, III, IV) is given after each statement (see section 3.4.6 for more information about the classification of evidence).
19 Here and elsewhere in the guideline, each study considered for review is referred to by a study ID (primary author and date of study publication, except where a study is in press or only submitted for publication, then a date is not used). 20 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
174 CONFIDENTIAL Second draft for consultation September 2003
Effect of treatment on remission from binge eating There is strong evidence suggesting that there is a clinically significant difference between CBT-BED and wait-list control with CBT-BED being superior in terms of remission (defined as cessation of binge eating) by the end of treatment (N = 4; n = 226; Random Effects RR = 0.64; 95% CI, 0.49 to 0.84; NNT = 3; 95% CI, 2 to 7). I
There is strong evidence suggesting that there is a clinically significant difference between simplified DBT and wait-list control with simplified DBT being superior in terms of remission (defined as cessation of binge eating) by the end of treatment (N = 1; n = 44; RR = 0.30; 95% CI, 0.15 to 0.60; NNT = 2; 95% CI, 1 to 3). I
There is strong evidence suggesting that there is a clinically significant difference between IPT-BED and wait-list control with IPT-BED being superior in terms of remission (defined as cessation of binge eating) by the end of treatment (N = 1; n = 38; RR = 0.56; 95% CI, 0.37 to 0.84; NNT = 3; 95% CI, 2 to 5). I
Effect of treatment on frequency of binge eating There is strong evidence suggesting that there is a clinically significant difference between CBT-BED and wait-list control with CBT-BED being superior on mean frequency of binge eating by the end of treatment (N = 4; n = 214; Random Effects SMD = -1.30; 95% CI, -2.13 to -0.48). I
There is insufficient evidence to determine whether simplified DBT has any impact on the frequency of binge eating by the end of treatment. I
There is strong evidence suggesting that there is a clinically significant difference between IPT-BED and wait-list control with IPT-BED being superior on mean frequency of binge eating by the end of treatment (N = 1; n = 38; SMD = -1.44; 95% CI, -2.16 to -0.72). I
There is limited evidence suggesting that there is a clinically significant difference between BWC and wait-list control with BWC being superior on mean frequency of binge eating by the end of treatment (N = 1; n = 82; SMD = -0.45; 95% CI, -0.89 to -0.01). I
Effect of treatment on weight There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and wait-list control in terms of mean body weight (BMI where possible) by the end of treatment (N = 3; n = 176; SMD = - 0.02; 95% CI, -0.33 to 0.30). I
There is insufficient evidence to determine whether BWC has any impact on body weight by the end of treatment. I
175 CONFIDENTIAL Second draft for consultation September 2003
Other effects of treatment There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and wait-list control in terms of mean depression scores by the end of treatment (N = 4; n = 214; SMD = -0.18; 95% CI, -0.46 to 0.10). I
There is limited evidence suggesting that there is a clinically significant difference between IPT-BED and wait-list control with IPT-BED being superior on mean depression scores by the end of treatment (N = 1; n = 38; SMD = -0.80; 95% CI, -1.46 to -0.13). I
There is insufficient or no evidence to determine whether CBT-BED, simplified DBT or IPT-BED have any impact on general psychiatric state or interpersonal/ social functioning by the end of treatment. I
Attrition from the study There is limited evidence suggesting that there is a clinically significant difference favouring the wait-list control group over CBT-BED in terms of the number of people leaving the study early due to any reason by end of treatment (N = 4; n = 222; RR = 1.86; 95% CI, 1.10 to 3.15; NNH = 7; 95% CI, 4 to 34). I
There is insufficient evidence to determine whether there is any difference between simplified DBT, IPT-BED or BWC and wait-list control in terms of the number of people leaving the study early due to any reason by the end of treatment. I
8.2.4 Psychological treatments compared with other psychological treatments The treatments included in this section are for BED only, usually in the presence of obesity.
8.2.4.1 Psychological treatments reviewed The following treatments were included: • Behavioural weight control (BWC)
• Cognitive behaviour therapy for binge eating disorder (CBT-BED)
• Guided self-help (GSH)
• Interpersonal psychotherapy for binge eating disorder (IPT-BED)
• Pure self-help (PSH).
8.2.4.2 Studies considered
176 CONFIDENTIAL Second draft for consultation September 2003
The review team conducted a new review of psychological treatments compared to other psychological treatments for BED. Four RCTs met the eligibility criteria set by the Psychological Topic Group (CARTER1998, LOEB2000, NAUTA2000, WILFLEY1993, WILFLEY2002). Of these, one compared CBT-BED with BWC (NAUTA2000) using a follow-up period of 6 months, and two compared CBT-BED with IPT-BED (WILFLEY1993, WILFLEY2002), with only the latter study using a follow-up of 12 months. A further two trials compared GSH with PSH (CARTER1998, LOEB2000), with the former study using a 6 month follow-up. Both studies used the book Overcoming Binge Eating (Fairburn, 1995).
Thus, four trials, involving a total of 404 participants, were included in this review.
Full details of studies included in this review and reasons for excluding studies are given in Appendix 17.
8.2.4.3 Evidence statements21
Effect of treatment on remission from binge eating There is insufficient evidence to determine whether or not there is any difference between CBT-BED and IPT-BED or BWC in terms of remission (defined as cessation of binge eating) by either the end of treatment or post- treatment follow-up. I
There is strong evidence suggesting that there is a clinically significant difference between CBT-BED and BWC with CBT-BED being superior in terms of remission (defined as cessation of binge eating) at follow-up (N = 1; n = 37; RR = 0.25; 95% CI, 0.08 to 0.79; NNT = 3; 95% CI, 2 to 8). I
There is insufficient evidence to determine whether GSH differs from PSH on remission of binge eating by end of treatment. I
Effect of treatment on frequency of binge eating It is unlikely that CBT-BED and IPT-BED differ with regard to their effect on mean frequency of binge eating by either the end of treatment or follow-up:
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and IPT-BED on mean frequency of binge eating by the end of treatment (N = 2; n = 194; SMD = -0.07; 95% CI, -0.35 to 0.22). I
21 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
177 CONFIDENTIAL Second draft for consultation September 2003
• There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and IPT-BED on mean frequency of binge eating at follow-up (N = 1; n = 138; SMD = 0.14; 95% CI, -0.19 to 0.48). I
There is insufficient evidence to determine whether CBT-BED differs from BWC on mean frequency of binge eating by post-treatment follow-up. I
There is limited evidence suggesting that there is a clinically significant difference between GSH and PSH with GSH being superior on mean frequency of binge eating by the end of treatment (N = 2; n = 109; SMD = - 0.48; 95% CI, -0.86 to -0.09). I
There is insufficient evidence to determine whether GSH differs from PSH on mean frequency of binge eating by post-treatment follow-up. I
Effect of treatment on weight There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and IPT-BED on mean weight (BMI where possible) by the end of treatment (N = 1; n = 158; SMD = 0.06; 95% CI, -0.26 to 0.37). I
There is insufficient evidence to determine whether CBT-BED differs from IPT-BED or BWC on mean weight by post-treatment follow-up. I
There is evidence suggesting that it is unlikely there is a clinically significant difference between GSH and PSH on mean weight (BMI where possible) by the end of treatment (N = 2; n = 109; SMD = 0.08; 95% CI, -0.30 to 0.46). I
There is insufficient evidence to determine whether GSH differs from PSH on mean weight by post-treatment follow-up. I
Other effects of treatment There is insufficient evidence to determine whether CBT-BED differs from IPT-BED or BWC in terms of depression by the end of treatment. I
There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and IPT-BED on mean depression scores at follow-up (N = 1; n = 138; SMD = 0.10; 95% CI, -0.24 to 0.43). I
There is insufficient evidence to determine whether CBT-BED differs from BWC in terms of depression by post-treatment follow-up. I
There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and IPT-BED on mean general psychiatric symptom scores by the end of treatment (N = 1; n = 158; SMD = 0.06; 95% CI, - 0.25 to 0.37). I
178 CONFIDENTIAL Second draft for consultation September 2003
There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and IPT-BED on mean psychosocial/ interpersonal functioning scores by the end of treatment (N = 2; n = 194; SMD = 0.06; 95% CI, -0.22 to 0.35). I
There is insufficient evidence to determine whether CBT-BED differs from IPT-BED in terms of mean psychosocial/ interpersonal functioning scores by post-treatment follow-up. I
There is evidence suggesting that it is unlikely there is a clinically significant difference between CBT-BED and IPT-BED on mean general psychiatric scores at follow-up (N = 1; n = 138; SMD = 0.13; 95% CI, -0.20 to 0.47). I
There is insufficient evidence to determine whether GSH differs from PSH in terms of mean general psychiatric scores by either the end of treatment or post-treatment follow-up. I
Acceptability of treatment There is insufficient evidence to determine whether CBT-BED is more, or less, acceptable to people with BED than IPT-BED or BWC. I
There is insufficient evidence to determine whether GSH is more, or less, acceptable to people with BED than PSH. I
8.2.5 Special considerations in the management of children and adolescents Nothing is known about the treatment of atypical eating disorders in adolescents. Binge eating disorder does occur in some children and adolescents, especially among those with obesity (e.g., Decaluwe et al., 2003). The prevalence of BED in this age group is not known. There have been no studies of their treatment. This omission needs to be rectified. In the meantime, it is the view of the GDG that child and adolescent patients with BED should receive the same type of treatment as adults but adapted to suit their age, circumstances and level of development and with appropriate family involvement.
8.2.6 Clinical summary There has been no research specifically directed at the treatment of atypical eating disorders other than BED. The view of the GDG is that clinicians should manage the large number of these cases according to the guidelines for anorexia nervosa or bulimia nervosa depending on the clinical presentation and age of the patient.
With regard to BED, given the apparently good response, at least in the short term, to a range of different psychological interventions including self-help
179 CONFIDENTIAL Second draft for consultation September 2003
and given the lower level of acute physical and psychiatric risk compared to anorexia and bulimia nervosa, treatment for BED may often be deliverable in primary care through the use of evidence-based self-help manuals. Community-based or clinical behavioural weight loss programmes may also help these patients, but only limited evidence is available. Improvements can be expected in terms of reduced frequency of binge eating, improved mood, increased remission from binge eating, but not in reduced body weight. Children and adolescents with binge eating problems should receive the same type of treatment as adults but adapted to suit their age, circumstances and level of development, with appropriate family involvement.
8.2.7 Clinical practice recommendations
8.2.7.1 In the absence of evidence to guide the management of atypical eating disorders (eating disorders not otherwise specified) other than binge eating disorder, it is recommended that the clinician considers following the guidance on the treatment of the eating problem that most closely resembles the individual patient’s eating disorder. (C)*
8.2.7.2 As a possible first step, patients with binge eating disorder should be encouraged to follow an evidence-based self-help programme. (B)
8.2.7.3 Healthcare professionals should consider providing direct encouragement and support to patients undertaking an evidence- based self-help programme as this may improve outcomes. This may be sufficient treatment for a limited subset of patients. (B)
8.2.7.4 Cognitive behaviour therapy for binge eating disorder (CBT-BED), a specifically adapted form of CBT, should be offered to adults with binge eating disorder. (A)*
8.2.7.5 Other psychological treatments (interpersonal psychotherapy for binge eating disorder and modified dialectical behaviour therapy) may be offered to adults with persistent binge eating disorder. (B)
8.2.7.6 Patients should be informed that all psychological treatments for binge eating disorder have a limited effect on body weight. (A)
8.2.7.7 When providing psychological treatments for patients with binge eating disorder consideration should be given to the provision of concurrent or consecutive interventions focusing on the management of any comorbid obesity. (C)
8.2.7.8 Suitably adapted psychological treatments should be offered to adolescents with persistent binge eating disorder. (C)
180 CONFIDENTIAL Second draft for consultation September 2003
8.3 Pharmacological interventions
8.3.1 Introduction Antidepressants, antiepileptics and appetite suppressants have been suggested as possible treatments for BED. The evidence base for such practice is very limited.
8.3.2 Current practice Little is known about the use of medication in the treatment of BED or other atypical eating disorders in the NHS.
8.3.3 Pharmacological treatment All the studies reviewed below are on BED.
8.3.3.1 Drugs reviewed The following drugs were included: • Antidepressants
• Antiepileptics.
Drugs that have had their licences withdrawn from the UK were not included in the guideline.
8.3.3.2 Studies considered for review The review team conducted a new review of RCTs involving pharmacological treatment in people with BED. Five trials met the eligibility criteria (ARNOLD2002, HUDSON1998, LAEDERACHHOFMAN1999, MCELROY2000, MCELROYINPRESS). Three trials compared an SSRI antidepressant (fluoxetine, fluvoxamine, sertraline) with placebo (ARNOLD2002, HUDSON1998, MCELROY2000), one trial compared a tricyclic antidepressant (imipramine) with placebo (LAEDERACHHOFMAN1999), and one trial (MCELROYINPRESS) compared an antiepileptic (topiramate) with placebo. Thus, five RCTs comparing a pharmacological treatment with placebo, involving 271 participants, were included for review.
Full details of studies included in this review and reasons for excluding studies are given in Appendix 17
8.3.3.3 Evidence statements22
Effect of treatment on remission
22 The full list of all evidence statements generated from meta-analyses (and the associated forest plots) will be available on the CD-ROM that accompanies the guideline. Where a meta- analysis was not possible (or not appropriate), a summary of the results from each study used to generate a statement can be found in Appendix 17.
181 CONFIDENTIAL Second draft for consultation September 2003
There is limited evidence suggesting that there is a clinically significant difference between antidepressants (SSRIs) and placebo with antidepressants being superior in terms of remission (defined as cessation of binge eating) by the end of treatment (N = 3; n = 179; RR = 0.77; 95% CI, 0.63 to 0.93; NNT = 6; 95% CI, 4 to 17). I
There is limited evidence suggesting that there is a clinically significant difference between topiramate and placebo with topiramate being superior in terms of remission (defined as cessation of binge eating) by the end of treatment (N = 1; n = 61; RR = 0.56; 95% CI, 0.34 to 0.92; NNT = 4; 95% CI, 2 to 15). I
Effect of treatment on symptoms There is limited evidence suggesting that there is a clinically significant difference between antidepressants and placebo with antidepressants being superior on mean frequency of binge eating by the end of treatment (N = 3; n = 91; SMD = -0.61; 95% CI, -1.04 to -0.18). I
Effect of treatment on weight There is insufficient evidence to determine whether antidepressants have any impact on weight when compared with placebo at the end of treatment. I
Other effects of treatment There is strong evidence suggesting that there is a clinically significant difference between antidepressants and placebo with antidepressants being superior on mean depression scores by the end of treatment (N = 2; n = 65; SMD = -0.78; 95% CI, -1.30 to -0.27). I
Acceptability of treatment There is insufficient evidence to determine whether antidepressants or antiepileptics are more, or less, acceptable to people with BED. I
Tolerability of treatment There is insufficient evidence to determine whether antidepressants or antiepileptics produce side effects in people with BED. I
8.3.4 Clinical summary There have been no studies of the use of drugs to treat atypical eating disorders other than BED. In BED there is limited evidence that by the end of treatment antidepressants can bring about improved remission from binge eating, reduced frequency of binge eating and improved mood. One small trial has demonstrated a positive impact of an antiepileptic (topiramate) on remission. No long-term follow-up data exists for any drug.
8.3.5 Clinical practice recommendations
8.3.5.1 As an alternative or additional first step to using an evidence-based self-help programme or a behavioural weight control programme,
182 CONFIDENTIAL Second draft for consultation September 2003
consideration should be given to offering a trial of a SSRI antidepressant drug to patients with binge eating disorder. (B)
8.3.5.2 Patients with binge eating disorders should be informed that SSRIs can reduce binge eating, but the long-term effects are unknown. Antidepressant drugs may be sufficient treatment for a limited subset of patients. (B)
8.4 Management of physical aspects The predominant long term risk for patients with BED are with the physical consequences of any comorbid obesity. As current treatment programmes for BED appear to have minimal or no impact on weight, appropriate physical monitoring, advice and management strategies for obesity should be adopted. The physical consequences and management of obesity is beyond the scope of this guideline.
Binge eating disorder is the predominant eating disorder to occur in Type 2 diabetes and this usually occurs in the context of obesity. In nearly 90% of cases the eating disorder preceded the onset of the diabetes (Herpertz et al., 1998).
The management of physical complications accompanying atypical eating disorders other than BED (e.g., severe underweight, electrolyte disturbance) should follow the guidance specified for anorexia nervosa and bulimia nervosa. 8.5 Service level interventions Little is known about the optimal management of people with atypical eating disorders (including BED) in the NHS. Sometimes the strict use of referral criteria specifying full syndrome eating disorders may mean that they are excluded from specialised services. Most patients with atypical eating disorders referred to secondary care are treated as outpatients unless the management of co-morbid states required admission to hospital. The GDG considered any evidence that certain levels of service provision – outpatient, day patient or inpatient – were associated with better outcomes in atypical eating disorders. No additional data relating specifically to atypical eating disorders (including BED) were identified. Therefore, it was agreed that the recommendations concerning the general approach for bulimia nervosa and anorexia nervosa be also applied to the atypical eating disorders. This should be borne in mind when considering the recommendations for this group of patients.
183 CONFIDENTIAL Second draft for consultation September 2003
9 Health economics evidence
Data on the economic burden of eating disorders and cost-effectiveness evidence of the different treatment options for eating disorders were collected and assessed as part of the development of the guideline in order to help decision-making.
9.1 Systematic literature review A systematic review of the health economic evidence in the field of eating disorders was conducted. The aim of the review was three-fold:
1) to identify all publications with information about the economic burden of eating disorders;
2) to identify previously conducted economic evaluations of any psychological, pharmacological or service level interventions for the treatment of eating disorders, or any interventions for the management of the physical aspects of eating disorders; and
3) to find studies with relevant resource use and cost data, or quality-of- life evidence generalisable to the UK context to facilitate possible cost- effectiveness modelling.
9.1.1 Search strategy Bibliographic electronic databases and health economic databases were searched for studies using the combination of a specially developed health economics search filter and a general filter for eating disorders. A combination of subject headings and free text searches were used where possible. The search strategies and the databases searched are presented in Appendix 12.
The search for further evidence included papers from reference lists of eligible studies and relevant reviews. Experts in the field of eating disorders and mental health economics were also contacted to identify additional relevant published and unpublished studies. Studies included in the clinical evidence review and stakeholders’ submissions were also screened for economic evidence.
9.1.2 Review process The database searches identified 770 possibly eligible references. A further five studies were identified by handsearching and recommendations from experts. Titles/abstracts of all references were checked to identify papers of
184 CONFIDENTIAL Second draft for consultation September 2003 potential relevance. The full texts of all potentially eligible studies (or where relevance/eligibility was not clear from the abstract (62 papers) were obtained and tested against a set of standard inclusion criteria by the health economist.
Papers eligible for inclusion were subsequently assessed for internal validity. The quality assessment of economic evaluations was based on the 32-point checklist used by the British Medical Journal to assist referees in appraisal of economic analyses (Drummond & Jefferson, 1996), or on a shortened 18-point version in the case of costing studies (Appendix 13).
9.1.3 Selection criteria
Cost-of-illness/ economic burden studies
• There was no restriction placed on language or publication status of the papers
• Studies published from 1980 until September 2002 were included. This date restriction was imposed in order to obtain data relevant to current health care settings and costs
• Only studies from OECD countries were included as the aim of the review was to identify economic burden information relevant to the current UK context
• Selection criteria based on types of clinical conditions and patients were identical to the clinical literature review section
• The study provided sufficient details regarding methods and results to enable judgement of the quality of the study and the use of the study’s data and results.
Economic evaluations
• The studies used an analytical method of cost-minimisation analysis, cost-effectiveness analysis, cost-utility analysis or cost-benefit analysis
• Clinical evidence was sourced from meta-analyses, randomised controlled trials, quasi-experimental trials or a cohort studies
• There was no restriction placed on language or publication status of the papers
185 CONFIDENTIAL Second draft for consultation September 2003
• Studies published from 1980 until September 2002 were included. This date restriction was imposed in order to obtain data relevant to current health care settings and costs
• Only studies from OECD countries were included as the aim of the review was to identify cost-effectiveness information relevant to the current UK context
• Selection criteria based on types of clinical conditions, patients, treatments and settings were identical to the clinical literature review section
• The studies provided sufficient details regarding methods and results to enable judgement of the quality of the study and the use of the study’s data and results.
Costing studies
• All types of costing studies were considered for inclusion, regardless of study design. However, this was subject to the study providing sufficient details on methods and results to enable judgement of the quality of the study
• There was no restriction placed on language or publication status of the papers
• Studies published from 1980 until September 2002 were included. This date restriction was imposed in order to obtain data relevant to current health care settings and costs
• Only studies from the UK were included as the aim of the review was to identify cost information relevant to the current UK context
• Selection criteria based on types of clinical conditions, patients, treatments and settings were identical to the clinical literature review section.
Health related quality-of-life studies
• All formal quality-of-life studies reporting utilities and involving people with eating disorders were considered for inclusion
• There was no restriction placed on language or publication status of the papers
186 CONFIDENTIAL Second draft for consultation September 2003
• Studies published from 1980 until September 2002 were included
• Selection criteria based on types of clinical conditions, patients, treatments and settings were identical to the clinical literature review section.
9.1.4 Data extraction Data were abstracted by a single abstractor using the economic data extraction form (Appendix 14). Masked assessment, whereby data extractors are blind to the details of journal, authors, etc., was not undertaken because there is no evidence to support the claim that this minimises bias (Cochrane, 2001).
9.1.5 Evidence synthesis
Cost-of-illness/economic burden studies Altogether 12 publications were deemed eligible for the economic burden review (Brown, 1997; Crow, 2003; Garvin, 2002; Hoek, 1991; Hoek, 2003; Howlett, 1995; Krauth, 2002; Lemouchoux, 2001; Office of Health Economics, 1994; Striegel-Moore, 2000; Turnbull, 1996; Vos, 2001). Results of these studies were summarised in the form of a narrative review about the socioeconomic burden of eating disorders in section 2.6 of the full guideline.
The GDG also had the initial attempt to establish the usual care pathway and health service use of people with eating disorders in the UK. However, the group later agreed that such task would require significant further primary research and so, it is outside the scope of the guideline development process.
Economic evaluations One economic evaluation was selected for data abstraction.
Koran et al. (1995) conducted an exploratory post hoc study that compared the cost-effectiveness of five different treatment strategies for 71 women with bulimia nervosa based on the randomised controlled trial by Agras et al. (1992, 1994). The five strategies were: 18 sessions of CBT (CB), 16 weeks and 24 weeks of desipramine (Med16, Med24) and CBT combined with desipramine for those durations (Combo16, Combo24). They calculated the average median treatment costs per successfully treated patient on an intention-to-treat basis. At 32 weeks these cost estimates were $3,948, $2,338, $2,972, $6,613 and $4,141, respectively. The same estimates at one-year were: $3,230, $3,117, $1,982, $6,613 and $4,832. The authors did not carry out sensitivity analysis. Overall, these estimates are average cost-effectiveness ratios, and so, cannot be used for the assessment of incremental efficiency.
187 CONFIDENTIAL Second draft for consultation September 2003
Costing studies Two costing studies relevant to the UK setting were included in the economic evidence review, although the quality of both studies was poor and they did not provide sufficient methodological details to be able to judge the robustness of their estimates.
A review by Meads et al. (2001) investigated the costs of outpatient and inpatient care for the treatment of anorexia nervosa in 1998 prices. They collected unpublished data and carried out a telephone survey of District Health Authorities in the West Midlands. The review showed that the mean cost per inpatient episode varied between £18,924 and £32,636. The mean number of outpatient sessions per patient per year varied between five and 13 with a cost range of £60-90 per outpatient episode. Based on these data the authors estimated that the cost of outpatient treatment is approximately 1/10 of the cost of inpatient treatment in the UK.
Birchall et al. (2002) costed an intensive day programme for the treatment of severe anorexia nervosa in Leicestershire. They compared the number of hospital days for their cohort of anorexia nervosa patients in the three years prior to the opening of the day programme (1994-1997) with another cohort in the three years following (1997-2000). They estimated the average cost of an inpatient day for eating disorder treatment at £216.73 and a day programme day at £111.07. The average number of inpatient days per patient greatly reduced after the introduction of the day programme from 217 to 90, but the number of day programme days significantly increased from 0 to 153. The average total cost of hospital use per patient was £3,919 and £2,434.20 before and after the day care programme was available, respectively.
Three further costing studies not relevant for the UK setting were also identified by the review (Kachele, 1999; Mitchell, 1998; Williamson, 2001).
Health related quality-of-life studies One study was identified that measured the quality-of-life of patients with eating disorders using the Nottingham Health Profile (NHP). Both patients with anorexia nervosa and bulimia nervosa showed significantly more impairment than average female student controls in the health domains of the NHP. In addition to health difficulties, the patient groups also reported functional difficulties in daily living. However, the study did not provide the necessary data to inform a cost-utility analysis (Keilen et al., 1994). 9.2 Cost-effectiveness modelling
9.2.1 Background The GDG in collaboration with the health economist discussed a few possible areas within the scope of the guideline with major cost impacts in an attempt to conduct primary economic evaluations alongside the guideline
188 CONFIDENTIAL Second draft for consultation September 2003 development process. Based on the preliminary analysis of the clinical evidence these areas were the following:
• Comparative cost-effectiveness of CBT versus IPT for the treatment of bulimia nervosa
• Comparative cost-effectiveness of individual CBT versus group CBT
• Comparative cost-effectiveness of psychotherapy versus antidepressant therapy for the treatment of bulimia nervosa
• Comparative cost-effectiveness of psychotherapy versus combination of psychotherapy and antidepressant therapy for the treatment of bulimia nervosa
• Comparative cost-effectiveness of antidepressant therapy versus combination of psychotherapy and antidepressant therapy for the treatment of bulimia nervosa
• Comparative cost-effectiveness of inpatient versus day patient management of anorexia nervosa
• Comparative cost-effectiveness of stepped care model versus conventional model for the management of eating disorders.
However, where there is little clinical evidence of comparative clinical efficacy/effectiveness of different treatment options, it is difficult to model the difference in cost-effectiveness between the alternatives.
The GDG identified one area where enough clinical data were available to develop a decision analytic model and provide robust cost-effectiveness information for the decision making process. This was the comparative cost- effectiveness of antidepressant therapy, cognitive behaviour therapy (CBT) and the combination of the two for the treatment of bulimia nervosa.
9.2.2 Treatment strategies and model structure As part of the health economic evidence presented for the guideline, a formal decision analytic model was constructed in order to explore the incremental cost-effectiveness of antidepressant therapy, individual cognitive behaviour therapy (CBT) and the combination of the two for the treatment of bulimia nervosa in the UK. Originally three strategies for the management of bulimia nervosa were modelled:
• Strategy A: antidepressant treatment given for 16 weeks
189 CONFIDENTIAL Second draft for consultation September 2003
• Strategy B: 20 sessions of CBT
• Strategy C: combination of 16 weeks antidepressant treatment and 20 sessions of CBT.
However, the clinical evidence reviewed in the guideline development process showed no overall superiority of combination therapy on treatment outcomes over CBT, and there was insufficient evidence to determine whether antidepressants differ from combination therapy with respect to the primary outcome by the end of treatment or post-treatment follow-up (see section 7.4). The clinical evidence together with the higher treatment cost of the combination therapy compared to the costs of either antidepressant therapy alone or CBT alone resulted in the dominance of the two single therapies over the combination therapy. Hence, strategy C was excluded from the final cost-effectiveness analysis.
Costs and outcomes of the alternative strategies were compared post- treatment. Although the original attempt was to compare the cost- effectiveness at 12-month follow-up, it was later amended according to the clinical evidence to four months, since there was insufficient evidence regarding the effects of antidepressant medication by post-treatment follow- up to draw firm conclusions about the relative outcome by the different treatment options (see section 7.4).
The structure of the decision tree is presented in Appendix 15.
9.2.3 Treatment outcomes Since no health related quality-of-life data of people with bulimia nervosa were available, economic evaluation in the form of a cost-utility analysis was not possible. In the clinical evidence review, more than one outcome measure was used; of these, remission/no remission from binge eating was chosen as the most appropriate primary outcome measure for the cost-effectiveness analysis.
Clinical parameter estimates were collected as part of the clinical evidence review. Mean values from the guideline meta-analyses were used as baseline estimates in the model, and 95% confidence intervals or minimum/maximum values from the clinical evidence reviews were used as the minimum/maximum estimates in the sensitivity analyses. Estimates of the absolute effectiveness of CBT were calculated from the relative risk data and the absolute effectiveness of antidepressant therapy to be as consequent as possible with the clinical evidence. No discounting of benefits was applied since the time horizon of the analysis did not exceed one year. Full details of the clinical outcome parameters used in this analysis are given in Appendix 16 and the corresponding clinical evidence review section of this guideline (Section 7.4).
190 CONFIDENTIAL Second draft for consultation September 2003
9.2.4 Resource use and unit costs Costing of the different treatment strategies was carried out as part of the guideline development process since no formal UK costing study of the different bulimia nervosa treatment strategies could be identified in the literature review.
The costs were identified from the UK National Health Service’s perspective using UK specific costs. All cost data are for year 2002/03. No discounting was applied since the time horizon of the analysis did not exceed one year.
Input data for the base case and the sensitivity analysis of the model were obtained from a broad range of sources. Resource utilization and unit cost data were collected as part of the literature review, from other published sources, or from the GDG acting as an expert panel. All baseline estimates, the ranges used in the sensitivity analyses and the sources of the information are listed in Appendix 16.
For calculating drug costs, the generic price of the drug from the British National Formulary 45 was used. However, the scenario of prescribing Prozac, the proprietary version of fluoxetine, instead of the generic equivalent was also explored in the sensitivity analysis.
Unit cost estimates for staff fees were taken from Netten et al. (2002). The staff unit cost estimates used in this analysis are without qualification costs, but include salary costs, salary oncosts, overheads, capital overheads and ongoing training costs.
Estimated resource utilization was then combined with the unit cost information to give the average cost associated with each treatment strategy. Treatment costs were also adjusted by the cost savings incurred by people leaving the treatment early.
No estimates were available for the total health service costs or for the health service use of an average patient with bulimia nervosa from the literature. This calculation was also outside of the scope of the guideline hence, no such formal estimate was available for the cost-effectiveness analysis.
9.2.5 Assumptions of the model
• A cohort of 1000 patients in each arm
• Each patient in the model has a well-established diagnosis of bulimia nervosa
• The treatment outcomes used are based on intention-to-treat analyses
191 CONFIDENTIAL Second draft for consultation September 2003
• CBT is provided by a suitably qualified and trained psychologist. (Clinical psychologist was chosen as a representative example of therapists providing bulimia nervosa specific CBT for the model.)
• The protocol of CBT used in the analysis is based on the clinical manual of bulimia nervosa specific CBT, (Fairburn et al, 1993) and personal communication with Prof. Christopher Fairburn
• The typical antidepressant treatment of bulimia nervosa is fluoxetine in 60mg/day dose (BNF, 2003)
• Since there was insufficient evidence that antidepressant therapy given for longer than 16 weeks is more effective than given only for 16 weeks, 16 weeks was chosen as the base case for the length of antidepressant therapy. However, the scenario of 24 weeks was also explored
• Antidepressant therapy can be prescribed in primary care by a general practitioner or in secondary care on an outpatient basis by a psychiatrist, and the clinical effectiveness of antidepressant does not depend on the qualification of the prescribing doctor
• The minimum amount of fluoxetine prescribed at one time is its monthly dose
• Those patients who leave the treatment early do not incur full treatment cost, only a proportion of it corresponding to the mean drop out time.
9.2.6 Result of the cost-effectiveness analysis
The patient groups to which the results apply are identical to those described in the clinical evidence section.
9.2.6.1 Treatment outcomes The systematically reviewed clinical evidence shows that the number of patients remitting is significantly higher for CBT than for antidepressant treatment. The end of treatment absolute risk of no remission by antidepressants was found to be 0.807 and the relative risk of no remission by antidepressant treatment versus CBT was 1.28 (see Paragraph 7.4). Although there was insufficient evidence to draw firm conclusions about the comparable longer-term treatment outcomes of CBT and antidepressant therapy, it is anticipated that the relapse rate with CBT is lower than that with antidepressants (Agras, 2001).
192 CONFIDENTIAL Second draft for consultation September 2003
9.2.6.2 CBT costs On average, one course of bulimia nervosa specific CBT costs £967.00 when provided by a suitably qualified and trained clinical psychologist.
The unit cost and resource utilization data used for this calculation are listed in Appendix 16.
9.2.6.3 Antidepressant treatment costs Due to the different service-level possibilities for prescribing antidepressant treatment for people with bulimia nervosa, multiple scenarios were considered to calculate the cost of fluoxetine therapy:
• The estimated average cost of generic fluoxetine treatment prescribed by a general practitioner is £118.48
• Fluoxetine prescribed by a psychiatrist in secondary care on an outpatient basis is estimated to be a less costly alternative of antidepressant therapy for people with bulimia nervosa, average total treatment cost of £94.66, than primary care provision when prescribed by a specialist registrar level physician. However, it is estimated to be more costly option, average total treatment cost of £238.66, when consultant level physician fees are used for the calculation.
No estimates exist for the health care costs due to the complications of antidepressant therapy in bulimia nervosa, and so they could not be included in the calculation.
9.2.6.4 Cost of no remission from bulimia nervosa Bulimia nervosa is a chronic psychiatric disorder, a high percentage of the people do not achieve remission at all or relapse in a few months post- treatment. Although no formal estimate exists about the magnitude of the additional health service use of people with bulimia nervosa, it is well known that people unsuccessfully treated continue to impose considerable extra costs for the health care sector (due to the need for extra eating disorder treatments, and additional medical and dental expenses due to symptomatic behaviour and co-morbidities). Patients with bulimia nervosa also incur substantial extra costs for the broader society due to lost productivity and have greatly decreased quality-of-life as shown by Keilen et al. (1994). Hence it is anticipated that CBT, which has a significantly higher remission rate compared to antidepressant treatment for people with bulimia nervosa, also averts important additional health care costs.
9.2.6.5 Incremental cost-effectiveness of CBT versus antidepressant therapy Since CBT was estimated to be both more effective and more costly, the difference in costs and effects were compared between CBT and
193 CONFIDENTIAL Second draft for consultation September 2003 antidepressant therapy. However, it needs to be emphasized that these estimates do not include the potential cost savings of CBT by averting additional and longer term health service use in the NHS. As a consequence, the net health service cost of CBT and the incremental cost-effectiveness ratio of CBT versus antidepressants are likely to be significantly overestimated in the analysis.
The incremental cost of CBT per successfully treated bulimia nervosa case is estimated to be the following:
• £4,807.24 when generic fluoxetine is prescribed by a general practitioner,
• £4,942.23/£4,126.41 when generic fluoxetine is prescribed by a psychiatrist in secondary care on an outpatient basis.
Comparison of the costs and effects of CBT and antidepressant therapy (AD) for two cohorts of 1000 people with bulimia nervosa are summarised in Table 1.
Table 1. Costs and effects of CBT versus antidepressant therapy for people with bulimia nervosa Incremental cost per successfully Incremental Incremental cost CBT AD treated bulimia effectiveness (£) nervosa case (CBT vs. AD) Number of people with bulimia nervosa 1000 1000 starting treatment Number of patients 370 193 177 achieving remission
Total treatment cost (GP £967,000 £118,483 £848,517 £4,807.24 prescription)
Total treatment cost (specialist £967,000 £94,656 £872,344 £4,942.23 registrar prescription)
Total treatment cost (consultant £967,000 £238,656 £728,344 £4,126.41 psychiatrist prescription)
194 CONFIDENTIAL Second draft for consultation September 2003
9.2.7 Sensitivity analyses
9.2.7.1 One-way sensitivity analysis There is considerable uncertainty about the parameter estimates used in the model and the policy implications of point estimates are uncertain. To explore the effect of this uncertainty a one-way sensitivity analysis was carried out, whereby individual parameters were varied while maintaining all remaining parameters at their base value. The ranges of parameters used in the sensitivity analyses together with the ranges of the calculated incremental cost-effectiveness ratios are listed in Table 2. The one-way sensitivity analysis shows that the most significant component of the uncertainty around the comparative cost-effectiveness of the two treatment strategies is the relative risk of no remission between antidepressant therapy and CBT. All other factors such as number of prescribing visits, length of prescribing visits, qualification of the prescribing physician, length of CBT sessions or the absolute risk of no remission by antidepressant therapy play only minor roles in the variation of the estimate.
Table 2. One-way sensitivity analysis Alternative value/range of Qualification of Alternative cost per successfully Parameter Base value prescribing physician value/range used treated bulimia nervosa case
Antidepressant £22.83 (generic £47.61 (Prozac) £4,357.99 cost/month fluoxetine)
Length of 16 weeks 24 weeks £4,471.61 antidepressant therapy
Number of antidepressant therapy 4 3-8 £4,752.02 - £4,517.70 visits
Length of General practitioner antidepressant therapy 9.36 minutes 4-15 minutes £4,954.62 - £4,652.16 visits
Length of CBT 50 minutes 40-60 minutes £3,711.54 - £5,902.94 sessions
Relative risk of no remission by 1.28 1.09-1.5 £12,735.84 - £3,154.75 antidepressants vs. CBT Absolute risk of no remission by 0.807 0.667-0.913 £5,818.42 - £4,248.37 antidepressants
£22.83 (generic Specialist Antidepressant cost £47.61 (Prozac) £4,492.98 fluoxetine)
Length of 16 weeks 24 weeks £4,674.10 antidepressant therapy
195 CONFIDENTIAL Second draft for consultation September 2003
Cost of specialist per £27 (specialist £207 (consultant £4,126.41 hour contact registrar) psychiatrist)
Number of antidepressant therapy 4 3-8 £4,861.70 - £4,804.56 visits
Length of antidepressant therapy 15 minutes 10-20 minutes £4,983.02 - £4,901.44 visits
Length of CBT 50 minutes 40-60 minutes £3,846.53 - £6,037.93 sessions
Relative risk of no remission by 1.28 1.09-1.5 £13,093.48 - £3,243.34 antidepressants vs. CBT Absolute risk of no remission by 0.807 0.667-0.913 £ 5,981.80 - £4,367.67 antidepressants
9.2.7.2 Worst-case/best-case scenario sensitivity analysis Worst-case/best-case scenarios were also investigated using the least favourable/most favourable parameter values for CBT and the most favourable/least favourable parameter values for antidepressant therapy, respectively. These showed that the extreme values of the incremental cost per successfully treated bulimia nervosa case are £19,209.12 and - £249.47.
9.2.7.3 Probabilistic sensitivity analysis To demonstrate the joint uncertainty around the parameters used in the cost- effectiveness model, a probabilistic sensitivity analysis was conducted. The theoretical basis of probabilistic sensitivity analysis is described in detail by Briggs and Gray (1999). The methodology applied in this analysis is based on the study by Briggs et al. (2002). Using the baseline estimates and the minimum/maximum values of the different variables, special distributions were assigned to all parameters included in the sensitivity analysis and then the incremental cost-effectiveness ratio was simulated 1000 times.
The result of the probabilistic analysis is illustrated in the form of a cost- effectiveness acceptability curve in Figure 1. This gives an estimate of the proportion of the simulated incremental cost-effectiveness ratios that lie below a given threshold. (The threshold value is the maximum value a decision maker is willing to pay for a unit of effect, in this case for an additional successfully treated bulimia nervosa case.) Equally, it shows the probability that CBT is cost-effective when compared to antidepressant treatment for bulimia nervosa in the UK.
196 CONFIDENTIAL Second draft for consultation September 2003
Cost-effectiveness acceptability curve
100.00%
90.00%
80.00%
70.00%
60.00%
50.00% GP Specialist registrar
40.00% Consultant psychiatrist
30.00% Probability that CBT is cost-effective 20.00%
10.00%
.00% 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 Decision-maker's willingness to pay threshold
Figure 1. Cost-effectiveness acceptability curves of CBT versus antidepressant therapy for the treatment of bulimia nervosa depending on the qualification of the prescribing physician
9.2.8 Discussion of the health economic evidence The systematic literature review identified only one economic evaluation within the scope of the guideline and two costing studies relevant to the UK setting, but no health related quality-of-life study with the necessary data for a cost-utility analysis. Based on the available literature, no firm conclusions could be made about the comparative cost-effectiveness of the different competing therapeutic options either for the treatment of anorexia nervosa, bulimia nervosa or EDNOS.
The question of the comparative cost-effectiveness of antidepressants and CBT for people with bulimia nervosa was identified within the scope of the guideline as having a major economic consequence and enough clinical evidence to conduct a primary cost-effectiveness analysis. Present analysis shows that CBT is both more effective and more costly than antidepressant therapy. The point estimate of its incremental cost per successfully treated bulimia nervosa case varies between £4,807.24, £4,942.23 and £4,126.41 depending on whether the antidepressant is prescribed in primary care or in secondary care by a more junior doctor or by a consultant psychiatrist, respectively. However, it is anticipated that these values are overestimations of the real incremental cost-effectiveness of CBT since its potential savings in additional health care costs were not included in the analysis due to the lack of available resource use data.
197 CONFIDENTIAL Second draft for consultation September 2003
Uncertainty around these estimates was explored by sensitivity analyses, including probabilistic analysis. Under present circumstances this shows that if decision makers are not willing to pay more for additional benefit, CBT is unlikely to be cost-effective. On the other hand, if decision makers are willing to pay £4,000 - £5,000 more for an additional successfully treated bulimia nervosa case, the probability of CBT being cost-effective is 50%. The likelihood of CBT being cost-effective would increase to approximately 95% if the decision maker’s willingness to pay threshold for the same benefit is £10,000. However, it is assumed that if broader health service costs could be included in the analysis, the cost-effectiveness curve would shift to the left and the probability of CBT being cost-effective would increase at each threshold values. Depending on the size of the additional averted health care costs by CBT, CBT could be even cost saving for the health service sector and clearly superior to antidepressant therapy.
Based on the reviewed evidence, combination therapy of CBT and antidepressants for the treatment of bulimia nervosa is highly unlikely to be cost-effective for the NHS.
Significant uncertainty around these results still exists (e.g. the true cost of side effects of antidepressant therapy are unknown, nor have attempts been made to quantify possible costs averted due to successful treatment). When further research is carried out, it will be necessary to re-estimate the cost- effectiveness of each alternative incorporating such influences. However, all these influences are likely to favour bulimia nervosa specific CBT and so, current estimates may be considered conservative.
No other cost-effectiveness analyses could be carried out as part of the guideline development process due to the lack of sufficient clinical evidence or available resource use data.
When such data will become available, other areas within the scope of this guideline identified as having possible major cost implications for the NHS should be explored (see Paragraph 9.2.1). Particularly the efficiency of a stepped care model for eating disorders is an important area considering the usual waiting time for and the great training requirements of psychotherapies. It is also recommended to investigate the comparative cost- effectiveness of IPT versus bulimia nervosa specific CBT for the treatment of bulimia nervosa in the UK with regard to the non-disease specific nature of IPT.
9.2.9 Implementation costs of the guideline There are insufficient data about the health service utilization patterns of patients with eating disorders and about the currently available health care resources for the treatment and management of eating disorders; hence at
198 CONFIDENTIAL Second draft for consultation September 2003 present it is impossible to calculate the estimated cost impact of the implementation of this guideline for the NHS.
However, it is anticipated that the recommended shift towards CBT in the management of bulimia nervosa would impose a great need for healthcare professionals trained in bulimia nervosa specific CBT. The NHS cost of bulimia nervosa specific CBT training per person was calculated using the resource use information provided by the GDG. The estimate is based on the teaching programme of the Department of Psychiatry, University of Oxford, UK including a two-day workshop, 20 four-hour long meetings and the additional time required by the trainer for a group of five trainees. Clinical psychologist was chosen as a representative example of trainees for bulimia nervosa specific CBT for the calculation. The total training cost per trainee was estimated to be £4,326 in year 2002/03 assuming that currently enough resources are available in the NHS to train healthcare professionals in bulimia nervosa specific CBT. Consequently, this estimate does not include any qualification costs. Further, possible travel costs and the cost of time spent on travelling related to the training could not be included in the analysis due to the lack of such data.
199 CONFIDENTIAL Second draft for consultation July 2003
10 Criteria for auditing the management of eating disorders
10.1 Objectives for the audit One or more audits can be carried out in different care settings to ensure that:
• Individuals with an eating disorder are involved in their care
• Treatment options, including psychological interventions, are appropriately offered for individuals with an eating disorder.
10.2 Individuals to be included in an audit A single audit could include all individuals with an eating disorder. Alternatively, individual audits could be undertaken on specific groups of individuals such as:
• People with a specific eating disorder e.g., bulimia nervosa
• Sample of patients from particular populations in primary care.
10.3 Measures that could be used as a basis for an audit See table below.
CONFIDENTIAL Second draft for consultation September 2003
Criterion Standard Exception Definition of terms
1. Cognitive behaviour therapy (CBT) in bulimia nervosa in adults
In patients with bulimia CBT should be offered to the Individuals who decline such The notes should indicate that nervosa, CBT specially great majority of individuals an offer of treatment, who the healthcare professional adapted for the disorder with bulimia nervosa assessed choose an alternative responsible has discussed the should be offered to adult as needing treatment in psychological intervention process and potential benefits patients assessed as needing secondary care services. (such as interpersonal of the intervention. treatment in secondary care psychotherapy) and those with The course of treatment should The notes should record if the services. severe comorbidity of a type normally be 16 to 20 individual patient completes a full course that will interfere with the The course of treatment should sessions provided by a trained of treatment. patient benefiting from CBT normally be 16 to 20 individual healthcare professional over 4 (for example, severe The course of the treatment sessions provided by a trained to 5 months. depression, marked substance should also be described in the healthcare professional over 4 abuse). notes and it should have to 5 months. followed the specific strategies
201 CONFIDENTIAL Second draft for consultation September 2003
and procedures employed in CBT-BN for bulimia nervosa (Fairburn et al., 1993).
2. Cognitive behaviour therapy (CBT) in bulimia nervosa in adolescents
Older adolescents with CBT should be offered to the Individuals who decline such The notes should indicate that bulimia nervosa may be majority of older adolescents an offer of treatment and those the healthcare professional treated with CBT-BN adapted with bulimia nervosa assessed with severe comorbidity or responsible has discussed the as needed to suit their age, as needing treatment in developmental problems of a process and potential benefits circumstances and level of secondary care services. type that will interfere with the of the intervention. development. patient benefiting from CBT. The course of treatment should The notes should record if the The course of treatment should normally be 16 to 20 individual patient completes a full course normally be 16 to 20 individual sessions provided by a trained of treatment. sessions over 4 to 5 months. healthcare professional over 4 The course of the treatment to 5 months. should also be described in the
202 CONFIDENTIAL Second draft for consultation September 2003
notes and it should have followed the specific strategies and procedures employed in CBT-BN for bulimia nervosa.
3. Psychological treatment in anorexia nervosa
Most adults with anorexia Psychological treatment Individuals who decline such The notes should indicate that nervosa should be managed on should be offered to the great an offer of treatment and those the healthcare professional an outpatient basis with majority of individuals with with severe comorbidity of a responsible has discussed the psychological treatment anorexia nervosa assessed as type that will interfere with the process and potential benefits provided by a health needing outpatient treatment patient benefiting from of the intervention. professional competent in the in secondary care services. psychological treatment (for The notes should record if the psychological treatment of example, severe depression, The course of treatment should patient completes a full course eating disorders. marked substance abuse). normally be for at least 6 of treatment. The course of treatment should months. The course of the treatment normally last for at least should also be described in the
203 CONFIDENTIAL Second draft for consultation September 2003
6 months. notes and it should have followed the specific strategies set out for the chosen intervention.
4. Family interventions in anorexia nervosa
For children and adolescents Family interventions focused Families who decline such an The notes should indicate that with anorexia nervosa, family explicitly on eating disorders offer of treatment, and the healthcare professional interventions that have an should be offered to the great possibly where the child or responsible has discussed the explicit focus on the eating majority of families with a adolescent is engaged in process and potential benefits disorder and family child or adolescent with individual psychological of the intervention. If the offer relationships should be offered anorexia nervosa who is not treatment. of intervention was not taken to the families of all patients in engaged in individual up, the notes should record secondary care services. psychological treatment. whether the parent, child or both declined the offer and the Both separate (parents and The course of treatment should number already in individual patient meet separately with normally last between 6 and 12
204 CONFIDENTIAL Second draft for consultation September 2003
the therapist) and conjoint months. psychological treatment. (parents and patient together The notes should record the with therapist) forms of family form of family intervention intervention may be beneficial. (separate or conjoint) and if the The course of treatment should family completed a full course normally last between 6 and of treatment. 12 months. The course of the treatment should also be described in the notes and it should have followed the specific strategies and procedures employed in family interventions for anorexia nervosa.
205 CONFIDENTIAL Second draft for consultation September 2003
5. Physical health review in anorexia nervosa
All patients with enduring Physical and mental health None. The notes should indicate that anorexia nervosa not under the review offered by GP to 100% the offer of a review was made care of secondary care services of patients with enduring to the patient and whether or should be offered an annual anorexia nervosa who are not not the patient attended for health review by their GP. in contact with secondary care review. services.
6. Inpatient care of anorexia nervosa
Patients with anorexia nervosa All patients requiring inpatient Individuals who are admitted An annual review of all who require admission to a care should be admitted to a as psychiatric emergencies to admissions for anorexia in psychiatric unit should be psychiatric unit experienced in general psychiatric wards. each PCT should be conducted admitted to a unit experienced the treatment of eating for all services that have in the treatment of eating disorders. provided inpatient services for disorders. anorexia nervosa.
206 CONFIDENTIAL Second draft for consultation September 2003
7. Patient satisfaction
All patients treated in All patients should be asked to Individuals who decline to The report should specify the secondary care for an eating complete a satisfaction complete the questionnaire. percentage of questionnaires disorder should be asked to questionnaire at the end of returned and the complete a satisfaction treatment. characteristics of those who questionnaire at the end of did and did not complete the The expected completion rate treatment. questionnaire. for the questionnaire is 50%.
8. Atypical eating disorders
In the absence of evidence to Patients with atypical eating None. The record system should guide the management of disorders are expected to record the diagnosis of all atypical eating disorders comprise at least 40% of patients assessed and taken on (eating disorders not otherwise patients assessed and taken on for treatment in a service. specified) other than as binge for treatment for eating eating disorder, it is disorders. recommended that the clinician considers following
207 CONFIDENTIAL Second draft for consultation September 2003
the guidance on treatment of the eating problem that most closely resembles the individual’s eating disorder.
208 CONFIDENTIAL Second draft for consultation July 2003
11 References
Agras, W. S., Dorian, B., Kirkley, B. G., Arnow, B., and Backman, J. Imipramine in the treatment of bulimia: a double-blind controlled study. International Journal of Eating Disorders 1987;6:29-38.
Agras, W. S., Schneider, J. A., Arnow, B., Raeburn, S. D., and Telch, C. F. Cognitive-behavioral and response -prevention treatments for bulimia nervosa. Journal of Consulting & Clinical Psychology 1989;57:215-221.
Agras, W. S., Rossiter, E. M., Arnow, B., Schneider, J. A., Telch, C. F., Raeburn, S. D., Bruce, B., Perl, M., and Koran, L. M. Pharmacologic and cognitive-behavioral treatment for bulimia nervosa: a controlled comparison. American Journal of Psychiatry 1992; 149:82-87.
Agras, W. S., Rossiter, E. M., Arnow, B., Telch, C. F., Raeburn, S. D., Bruce, B., and Koran, L. M. One-year follow-up of psychosocial and pharmacologic treatment for bulimia nervosa. Journal of Clinical Psychiatry 1994; 55:179-183.
Agras, W. S., Telch, C. F., Arnow, B., Eldredge, K. L., Detzer, M. J., Henderson, J., and Marnell, M. Does interpersonal therapy help patients with binge eating disorder who fail to respond to cognitive-behavioral therapy? Journal of Consulting & Clinical Psychology 1995; 63:356-360.
Agras S, Hammer L, McNicholas F..,A prospective study of the influence of eating-disordered mothers on their children.Int J Eat Disord. 1999 Apr;25(3):253-62.
Agras, W. S., Walsh, B. T., Fairburn, C. G., Wilson, C. T., and Kramer, H. C. A multicenter comparison of Cognitive-behavioral therapy and Interpersonal Psychotherapy for bulimia nervosa. Archives of General Psychiatry 2000a; 54:459-465.
Agras et al., 2000b Agras WS, Crow SJ, Halmi KA, Mitchell JE, Wilson GT, Kraemer HC. Outcome predictors for the cognitive behavior treatment of bulimia nervosa: data from a multisite study. Am J Psychiatry. 2000 Aug;157(8):1302-8.
Agras, W.S. The consequences and costs of the eating disorders. 2001; 24(2):371-379.
American Psychiatric Association Work Group on Eating Disorders. Practice guideline for the treatment of patients with eating disorders (revision). American Journal of Psychiatry 157, 1- 39. 2000.
APA 1994 American Psychiatric Association Work Group on Eating Disorders. Practice guideline for the treatment of patients with eating disorders (revision). American Journal of Psychiatry, 2000; 157(1 Suppl): 1-39.
Anstine D, and Grinenko D. Rapid screening for disordered eating in college –aged females in the primary care setting. Journal of Adolescent Health 2000; 26 (5):338-342.
Anstine et al., 1999
Arii, I., Yamashita, T., Kinoshita, M., Shimizu, H., Nakamura, M., and Nakajima, T. Treatment for inpatients with anorexia nervosa: Comparison of liquid formula with regular meals for improvement from emaciation. Psychiatry & Clinical Neurosciences 1996; 50:55-59. CONFIDENTIAL Second draft for consultation September 2003 Arnold, L. M., McElroy, S., Hudson, J. I., Welge, J. A., Bennett, A. J., and Keck, P. E. A placebo-controlled, randomized trial of fluoxetine in the treatment of binge-eating disorder. Journal of Clinical Psychiatry 2002; 63:1028-1033.
Attia, E., Haiman, C., Walsh, B. T., and Flater, S. R. Does Fluoxetine Augment the Inpatient Treatment of Anorexia Nervosa? American Journal of Psychiatry 1998; 155:548-551.
Babyak M, et al: Exercise treatment for major depression: maintenance of therapeutic benefit at 10 months. Psychosomatic Medicine 2000; 62:633-638.
Bachar, E., Latzer, Y., Kreitler, S., and Berry, E. M. Empirical comparison of two psychological therapies. Journal of Psychotherapy Practise and Research 1999; 8:115-128.
Bachrach LK, Guido D, Katzman D, Litt IF, Marcus R: Decreased bone density in adolescent girls with anorexia nervosa. Pediatrics 1990; 86:440-447.
Bailer, U., de Zwaan, M., Leisch, F., Strnad, A., Lennkh-Wolfsberg, C., El-Giamal, N., Hornik, K., and Kasper, S. Guided self-help versus cognitive behavioural group therapy in the treatment of bulimia nervosa. British Journal of Psychiatry 2002; In Press.
Beglin SJ, Fairburn CG What is meant by the term "binge"?Am J Psychiatry. 1992 Jan;149(1):123-4.
Ben-Tovim DI, Walker K, Gilchrist P, Freeman R, Kalucy R, Esterman A Outcome in patients with eating disorders: a 5-year study. Lancet 2001 Apr 21; 357(9264):1254-7
Bergh, C., Brodin, U., Lindberg, G., and Sodersten, P. Randomized controlled trial of a treatment for anorexia and bulimia nervosa. Proceeding of the National Academy of Sciences of the United States of America 2002; 99:9486-9491.
Beumont PJ, Abraham S.F, Argall W.J., George C.W., Glaun D.E: The onset of anorexia nervosa. Australian & New Zealand Journal of Psychiatry 1978; 12:145-149.
Beumont, P. J. V., Russell, J. D., Touyz, S. W., Buckley, C., Lowinger, K., Talbot, P., and Johnson, G. F. S. Intensive nutritional counselling in bulimia nervosa: a role for supplementation with fluoxetine? Australian & New Zealand Journal of Psychiatry 1997; 31:514-524.
Bianchi AS, Toy EC, Baker IB. The evaluation of involuntary weight loss. Primary Care Update for Ob/Gyns 1998; 5(6):263-267.
Biederman, J., Herzog, D., Rivinus, T. M., Harper, G. P., Ferber, R. A., Rosenbaum, J. F., Harmatz, J. S., Tondorf, R., Orsulak, P. J., and Schildkraut, J. J. Amitriptyline in the treatment of anorexia nervosa: a double-blind, placebo-controlled study. Journal of Clinical Psychopharmacology 1985; 5:10-16.
Biller BM, Saxe V, Herzog DB, Rosenthal DI, Holzman S, Klibanski A: Mechanisms of osteoporosis in adult and adolescent women with anorexia nervosa. Journal of Clinical Endocrinology and Metabolism 1989; 68:548-554.
Birchall, H.; Palmer, R.L.; Waine, J.; Gadsby, K.; Gatward, N. Intensive day programme treatment for severe anorexia nervosa- the Leicester experience. Psychiatric Bulletin, 2002; 26:334-336.
Birmingham, C. L., Goldner, E. M., and Bakan, R. Controlled trial of zinc supplementation in anorexia nervosa. International Journal of Eating Disorders 1994; 15:251-255.
210 CONFIDENTIAL Second draft for consultation September 2003 Bizeul C, Sadowsky N, Rigaud D. The prognostic value of initial EDI scores in anorexia nervosa patients: a prospective follow-up study of 5--10 years. Eating Disorder Inventory. European Psychiatry. 2001; 16(4): 232-8.
Briggs, A.H.; Goeree, R.; Blackhouse, G.; O’Brien, B.J. Probabilistic analysis of cost- effectiveness models: Choosing between treatment strategies for gastroesophageal reflux disease. Medical Decision Making, 2002; July-Aug: 290-308.
Briggs, A.H.; Gray, A.M. Handling uncertainty when performing economic evaluation of healthcare interventions. NICE Health Technology Assessment, 1999; 3(2).
Brinch M, Isager T, Tolstrup K: Anorexia nervosa and motherhood: reproduction pattern and mothering behavior of 50 women. Acta Psychiatrica Scandinavica 1988; 77:611-617.
Brinch, M., Isager, T., and Tolstrup, K. Patients' evaluation of their former treatment for anorexia nervosa (AN). Nordisk psykiatrisk tidsskrift. Nordic journal of psychiatry. 1988; 42:445-448. [BOTH REFERENCES 1988]
British National Formulary (BNF). 2003; 45, British Medical Association and the Royal Pharmaceutical Society of Great Britain.
Brown A: Eating disorders- the developing mixed economy. Mental Health Research Review 1997; 4:32-35.
Braun, DL, Sunday, SR, Halmi, KA. Psychiatric comorbidity in patients with eating disorders. Psychological Medicine 1994; 24: 859-67.
Bryant-Waugh R, Knibbs J, Fosson A, Kaminski Z, Lask B. Long term follow up of patients with early onset anorexia nervosa. Archives of Disease in Childhood 1988; 63(1):5-9.
Bryant-Waugh R.: Overview of the eating disorders, Anorexia Nervosa and Related Eating Disorders in Childhood and Adolescence Hove. Edited by Lask B, Bryant-Waugh R. Psychology Press, 2000, pp 27-40.
Bryant-Waugh RJ, Lask BD 1995 Eating disorders in children.J Child Psychol Psychiatry. 1995 Feb;36(2):191-202.
Bryant-Waugh RJ, Lask BD, Shafran RL, Fosson AR. Do doctors recognise eating disorders in children? Archives of Disease in Childhood 1992; 67(1):103-105.
Bulik CM, Sullivan PF, Carter FA, McIntosh VV, Joyce PR. The role of exposure with response prevention in the cognitive-behavioural therapy for bulimia nervosa. Psychol Med. 1998 May;28(3):611-23.
Bulik CM, Sullivan PF, Fear JL, Pickering A, Dawn A, McCullin M: Fertility and reproduction in women with anorexia nervosa: a controlled study. Journal of Clinical Psychiatry 1999; 60:130-135.
Bulik C.M., Sullivan P.F., Wade T. & Kendler K.S. (2000) Twin studies of eating disorders: a review. International Journal of Eating Disorders, 27, 1-20.
Bulik CM, Sullivan PF, Fear JL, Pickering A. Outcome of anorexia nervosa: eating attitudes, personality, and parental bonding. International Journal of Eating Disorders 2002; 28(2):139- 47.
211 CONFIDENTIAL Second draft for consultation September 2003 Burns T, Crisp AH. Outcome of anorexia nervosa in males. British Journal of Psychiatry 1984 Sep; 145:319-325.
Buston, K. Adolescents with mental health problems: what do they say about health services? Journal of Adolescence 2002; 25:231-242.
Carney CP, Andersen AE. Eating disorders. Guide to medical evaluation and complications. Psychiatric Clinics of North America 1996; 19(4):657-679.
Carruba, M. O., Cuzzolaro, M., Riva, L., Bosello, O., Liberti, S., Castra, R., Dalle, Grave R., Santonastaso, P., Garosi, V., and Nisoli, E. Efficacy and tolerability of moclobemide in bulimia nervosa: a placebo-controlled trial. International Clinical Psychopharmacology 2001;16:27-32.
Carter and Moss 1984 Screening for anorexia and bulimia nervosa in a college population: problems and limitations. Addict Behav. 1984;9(4):417-9.
Carter & Bulik 1994Carter, J. and Fairburn, C. C. Cognitive behavioral self-help for binge eating disorder: A controlled effectiveness study. Journal of Consulting & Clinical Psychology 1998;66:616-623.
Carter, J., Olmsted, M., Kaplan, A. S., McCabe, R. E., Mills, J. S., and Aime, A. Self-help for bulimia nervosa: a randomized controlled trial. American Journal of Psychiatry 2003; 160: 973-978.
Casper RC. Introduction to special issue. Journal of Youth and Adolescence 1996; 25:413-418.
Casper, Regina C; Jabine, L. N. An eight-year follow-up: Outcome from adolescent compared to adult onset anorexia nervosa Journal of Youth & Adolescence 1996; 25(4):499-517.
Castro J, Lazaro L, Pons F, Halperin I, Toro J: Adolescent anorexia nervosa: the catch-up effect in bone mineral density after recovery. Journal of the American Academy of Child and Adolescent Psychiatry 2001; 40:1215-1221.Channon, S., de Silva, P., Hemsley, D., and Perkins, R. A controlled trial of cognitive-behavioural and behavioural treatment of anorexia nervosa. Behavioral Research and Therapy 1989; 27:529-535.
Chen, E., Touyz, S. W., Beumont, P. J., Fairburn, C. G., Griffiths, R., Butow, P., Russell, J., Schotte, D. E., Gertler, R., and Basten, C. Comparison of group and individual cognitive- behavioural therapy for patients with bulimia nervosa. International Journal of Eating Disorders 2003; 33:241-254.
Cole et al., 1995
Cole, T. J., Bellizzi, M. C., Flegal, K. M., et al., (2000) Establishing a standard definition for child overweight and obesity worldwide: international survey British Medical Journal; 320:1240-1243.
Cooper & Fairburn, 1987
Cooper, P. J. and Steere, J. A comparison of two psychological treatments for bulimia nervosa: Implications for models of maintenance Behaviour Research and Therapy 1995; 33:875-885.
Cooper PJ, Coker S, Fleming C. An evaluation of the efficacy of supervised cognitive behavioral self-help bulimia nervosaJ Psychosom Res. 1996 Mar;40(3):281-7.
212 CONFIDENTIAL Second draft for consultation September 2003 Cooper PJ, Watkins B, Bryant-Waugh R, Lask B. The nosological status of early onset anorexia nervosa. Psychol Med. 2002 Jul;32(5):873-80.
Cotton MA, Ball C and Robinson P. Four simple questions can help screen for eating disorders. Journal of General Internal Medicine 2003
Crisp A H, Callender J S, Halek C, Hsu L K G: Long-Term Mortality in Anorexia Nervosa: A 20 Year Follow-Up of the St George's and Aberdeen Cohorts. British Journal of Psychiatry 1992; 161:80-87.
Crisp AH, Hsu LK, Stonehill E. Personality, body weight and ultimate outcome in anorexia nervosa. Journal of Clinical Psychiatry 1979; 40(8):332-325.
Crisp, A. H., Norton, K., Gowers, S., Halek, C., Bowyer, C., Yeldham, D., Levett, G., and Bhat, A. A controlled study of the effect of therapies aimed at adolescent and family psychopathology in anorexia nervosa British Journal of Psychiatry 1991; 59:325-333.
Crisp (1995) The dyslipophobias: a view of the psychopathologies involved and the hazards of construing anorexia nervosa and bulimia nervosa as 'eating disorders'. Proc Nutr Soc. 1995 Nov;54(3):701-9. Review.
Crisp A.H (2000) Stigmatisation of people with mental illnesses. British Journal of Psychiatry, 177, 4-7.Crow S, Praus B, Thuras P: Mortality from Eating Disorders - A 5-10 Year Record Linkage Study. International Journal of Eating Disorders 1999; 26:97-101.
Crow SJ, Peterson CB: The economic and social burden of eating disorders: A review, Eating disorders. Edited by Maj M, et al. John Wiley & Sons Ltd, 2003.
Dally PJ Anorexia Nervosa. London, Heinemann Medical Books 1969.[reference added to text]
Danziger Y, Mukamel M, Zeharia A, Dinari G, Mimouni M: Stunting of Growth in Anorexia Nervosa During the Prepubertal and Pubertal Period. Israeli Journal of Medical Science 1994; 30:581-584.
Dare, C., and Crowther, C. Living dangerously: psychoanalytic psychotherapy of anorexia nervosa. In Eating Disorders: Handbook of Theory, Treatment and Research (eds G. Szmukler, C. Dare & J. Treasure). Chichester: John Wiley & Sons. 1995.
Dare, C., Eisler, I., Russell, G., Treasure, J., and Dodge, L. Psychological therapies for adults with anorexia nervosa: randomised controlled trial of out-patient treatments British Journal of Psychiatry 2001; 178:216-221.
Deering S. Eating disorders: Recognition, evaluation, and implications for obstetrician/gynecologists. Primary Care Update for Ob/Gyns 2000; 8(1):31-35.
Department of Health (1996)
Department of Health (1991)
Department of Health (2002a)
Department of Health (2002b)
213 CONFIDENTIAL Second draft for consultation September 2003 Department of Health. A commitment to carers; Help for relatives, partners and friends of people with a mental health problem 2003. http://www.rethink.org/carers/.
Deter et al., 1990
Deter HC, Herzog W, Petzold E. The Heidelberg-Mannheim study: long-term follow-up of anorexia nervosa patients at the university medical center - background and preliminary results. In Herzog W et al. (eds.) The course of eating disorders: long-term follow-up studies of anorexia and bulimia nervosa. 1992; Berlin Springer pp71-84.
Deter HC, Herzog W. Anorexia nervosa in a long-term perspective: results of the Heidelberg- Mannheim Study. Psychosomatic Medicine 1994; 56(1):20-27.
Deter HC, Kopp W, Zipfel S, Herzog W. Male anorexia nervosa patients in long-term follow- up Nervenarzt. 1998; 69(5):419-426.
Deter et al., 1989 [check not reference above – date incorrect.]
Dolan RJ, Mitchell J, Wakeling A: Structural brain changes in patients with anorexia nervosa. Psychological Medicine 1988; 18:349-353.
Eating disorders: a guide for the primary care physician. Primary Care; Clinics in Office Practice 2002; 29(1):81-98.
Eckert E D, Halmi K A, Marchi P, Grove W, Crosby R: Ten Year Follow-Up of Anorexia Nervosa: Clinical Course and Outcome. Psychological Medicine 1995; 25:143-156.
Eckert ED, Goldberg SC, Halmi KA, Casper RC & Davis JM 1979 Behaviour therapy in Anorexia Nervosa British Journal of Psychiatry 134; 55-59.
Eddy KT, Keel PK, Dorer DJ, Delinsky SS, Franko DL, Herzog DB. Longitudinal comparison of anorexia nervosa subtypes. International Journal of Eating Disorders 2002; 31(2):191-201.
Eiro, M., Katoh, T., & Watanabe, T. Use of a proton-pump inhibitor for metabolic disturbances associated with anorexia nervosa. New England Journal of Medicine 2002; 346 (2), 140.
Eisler, I., Dare, C., Hodes, M., Russell, G., Dodge, E., and Le Grange, D. Family therapy for adolescent anorexia nervosa: the results of a controlled comparison of two family interventions. Journal of Child Psychology & Psychiatry & Allied Disciplines 2000; 41:727-736.
Eisler et al., 2003
Eldredge, K. L., Agras, W. S., Arnow, B., Telch, C. F., Bell, S., Castonguay, L., and Marnell, M. The effects of extending cognitive-behavioural therapy for binge eating disorder among initial treatment nonresponders. International Journal of Eating Disorders 1997; 21:347-352.
Fairburn, 1981 A cognitive behavioural approach to the treatment of bulimia.Psychol Med. 1981 Nov;11(4):707-11.
Fairburn, C. G., Marcus, M.D., & Wilson, G. T. (1993). Cognitive-behavioral therapy for binge eating and bulimia nervosa: A comprehensive treatment manual. In C. G. Fairburn & G. T. Wilson (Eds.), Binge eating: Nature, assessment and treatment (pp. 361-404). New York: Guilford Press.
Fairburn, 1995, Overcoming Binge Eating disorders. Guilford Press.
214 CONFIDENTIAL Second draft for consultation September 2003 Fairburn, 1997
Fairburn C.G., Cooper Z., Doll H.A., Norman P., O'Connor M.: The natural course of bulimia nervosa and binge eating disorder in young women. Archives of General Psychiatry 2000; 57:659-665.
Fairburn C.G., Harrison P.J.: Eating Disorders. The Lancet 2003; 361:407-416.
Fairburn CG, Brownell KD: Eating disorders and obesity. Guilford Press, 2001.
Fairburn & Peveler, 1990 Bulimia nervosa and a stepped care approach to management. Gut. 1990 Nov;31(11):1220-2. Review.
Fairburn & Cooper, 1984 Binge-eating, self-induced vomiting and laxative abuse: a community study. Psychol Med. 1984 May;14(2):401-10.
Fairburn & Cooper, 1987
Fairburn & Cooper, 1982 Fairburn CG, Cooper PJ. Self-induced vomiting and bulimia nervosa: an undetected problem. Br Med J (Clin Res Ed). 1982 Apr 17;284(6323):1153-5.
Fairburn & Cooper, 1993
Fairburn et al., 1996 Fairburn CG, Welch SL, Norman PA, O'Connor ME, Doll HA.Bias and bulimia nervosa: how typical are clinic cases? Am J Psychiatry. 1996 Mar;153(3):386-91.
Fairburn & Beglin, 1990 Studies of the epidemiology of bulimia nervosa. Am J Psychiatry. 1990 Apr;147(4):401-8.
Fairburn, Beglin & Davies, 1992
Fairburn, Hay & Walsh, 1993
Fairburn, C. G., Jones, R., Peveler, R., and Hope, R. A. Psychotherapy and bulimia nervosa. Longer-term effects of interpersonal psychotherapy, behavior therapy, and cognitive behavior therapy. Archives of General Psychiatry 1993;50:419-428.
Fairburn, C. G., Jones, R., Peveler, R., Carr, S. J., Solomon, R. A., O'Conner, M., Burton, J., and Hope, R. A. Three psychological treatments for bulimia nervosa: A comparative trial. Archives of General Psychiatry 1991; 48:463-469.
Fairburn, C. G., Jones, R., Peveler, R., Hope, T., and O'Connor, M. Predictors of 12 month outcome in bulimia nervosa and the influence of attitudes to shape and weight. Journal of Consulting & Clinical Psychology 1993; 61:696-698.
Fairburn, C. G., Kirk, J., O'Conner, M., and Cooper, P. J. A comparison of two psychological treatments for bulimia nervosa. Behaviour Research and Therapy 1986;24:629-643.
Fairburn, C. G., Norman, P. A., Welch, S. L., O'Connor, M., Doll, H. A., and Peveler, R. A prospective study of outcome in bulimia nervosa and the long term effects of three psychological treatments. Archives of General Psychiatry 1995;52:304-312.
Fairburn C.G., Cowen P.J. & Harrison P.J. (1999) Twin studies and the aetiology of eating disorders. International Journal of Eating Disorders, 26, 349-358.
215 CONFIDENTIAL Second draft for consultation September 2003
Fairburn C.G., Cooper Z., Doll H., Norman P. & O’Connor M. (2000) The natural course of bulimia nervosa and binge eating disorder in young women. Archives of General Psychiatry, 57, 659-665.
Fairburn CG, Harrison PJ. Eating disorders. Lancet. 2003 Feb 1;361(9355):407-16.
Faris, P. L., Meller, W. H., Goodale, R. L., Oakman, S. A., Hofbauer, R. D., Marshall, A. M., Daughters, R. S., Banerjee-Stevens, D., Eckert, E. D., and Hartman, B. K. 5-HT3 antagonist therapy of bulimia nervosa: a peripherally active agent for a central nervous system eating disorder? Gastroenterology 2000; 119:271-272.
Faris, P. L., Won Kim, S., Meller, W. H., Goodale, R. L., Oakman, S. A., Hofbauer, R. D., Marshall, A. M., Daughters, R. S., Banerjee-Stevens, D., Eckert, E. D., and Hartman, B. K. Effect of decreasing afferent vagal activity with ondansetron on symptoms of bulimia nervosa: a randomised, double-blind trial. The Lancet 2000; 355:792-797.
Fassino, S., Leombruni, P., Daga, G., Brustolin, A., Migliaretti, G., Cavallo, F., Rovera, G. Efficacy of citalopram in anorexia nervosa: a pilot study. European Neuropsychopharmacology 2002; 12(5):453-459.
Fichter MM, Quadflieg N. Six-year course and outcome of anorexia nervosa. International Journal of Eating Disorders. 1999 Dec; 26(4):359-85.
Fichter, M. M., Leible, K., Rief, W., Brunner, E., Schmidt-Auberger, S., and Engel, R. R. Fluoxetine versus placebo: a double blind study with bulimic inpatients undergoing intensive psychotherapy. Pharmacopsychiatry 1991;24:1-247.
Fichter, M., Kruger, R., Rief, W., Holland, R., and Dohne, J. Fluvoxamine in prevention of relapse in bulimia nervosa: effects on eating-specific psychopathology. Journal of Clinical Psychopharmacology 1996;16:9-18.
Field, T., Schanberg, S., Kuhn, C., Field, T., Fierro, K., Henteleff, T., Mueller, C., Yando, R., Shaw, S., and Burman, I. Bulimic adolescents benefit from massage therapy. Adolescence 1998; 33, 555-563.
Fluoxentine for anorexia nervosa. Nurses Drug Alert 2001; 25:47-48.
Fluoxetine Bulimia Nervosa Collaborative Study Group. Fluoxetine in the treatment of bulimia nervosa. A multicenter, placebo-controlled, double-blind trial. Archives of General Psychiatry 1992; 49:139-147.
Franko DL, Blais MA, Becker AE, Delinsky SS, Greenwood DN, Flores AT, Eddy KT, Herzog DB: Pregnancy complications and neonatal outcomes in women with eating disorders. American Journal of Psychiatry 2001; 158:1461-1466.
Freeman, C., Barry, F., Dunkeld-Turnbull, J., and Henderson. Controlled trial of psychotherapy for bulimia nervosa. British Medical Journal 1988; 296:521-525.
Freeman, C., Sinclair, F., Turnball, J., and Annandale, A. Psychotherapy for bulimia : A controlled study. Journal of Psychiatric Research 1985; 19:473-478.
Freund, KM, Graham SM, Lesky LG and Moskovitz MA. Detection of bulimia nervosa in a primary care setting. Journal of General Internal Medicine 1993; 8(5):236-242.
216 CONFIDENTIAL Second draft for consultation September 2003 Garner and Garfinkel The Eating Attitudes Test: an index of the symptoms of anorexia nervosa. Psychol Med. 1979 May;9(2):273-9.
Garner el al 19835. Garner DM, Olmsted MP, Polivy J. Development and validation of a multidimensional Eating Disorders Inventory for anorexia nervosa and bulimia. Int J Eat Dis 1983;2:15-34.
Garner, D. M., Rockert, W., Davis, R., Garner, M. V., Olmsted, M., and Eagle, M. Comparison of cognitive-behavioural and supportive-expressive therapy for bulimia nervosa. American Journal of Psychiatry 1993; 150:37-46.
Garvin, V.; Striegel-Moore, R.H. Health services research for eating disorders in the United States: A status report and a call to action. In: Striegel-Moore, R.H. et al. (eds.) Eating disorders: Innovative directions in research and practice. 2001; APA, Washington D.C.
Geist, R., Heinmaa, M., Stephens, D., Davis, R., and Katzman, D. K. Comparison of family therapy and family group psychoeducation in adolescents with anorexia nervosa. Canadian Journal of Psychiatry - Revue Canadienne de Psychiatrie 2000;45:173-178.
Goldberg SC, Eckert ED, Casper RC, Halmi KA, Davis JM, Roper MT. Factors influencing hospital differences in weight gain in anorexia nervosa. Journal of Nervous and Mental Disease. 1980; 168(3):181-183.
Goldberg, S. C., Casper, R. C., and Eckert, E. D. Effects of cyproheptadine in anorexia nervosa. Psychopharmacology Bulletin 1980;16:29-30.
Goldberg, S. C., Halmi, K. A., Eckert, E. D., Casper, R. C., and Davis, J. M. Cyproheptadine in anorexia nervosa. British Journal of Psychiatry 1979; 134:67-70.
Goldbloom, D. S., Olmsted, M. P., Davies, R., and Shaw, B. A randomized control trial of fluoxetine and cognitive behavioural therapy for bulimia nervosa: short-term outcome. Behavioral Research and Therapy 1997; 35:803-811.Goldbloom, D. S. & Kennedy, S. H. (1995). Medical complications of anorexia nervosa. In K. D. Brownell & C. G. Fairburn (Eds.), Eating disorders and obesity: A comprehensive handbook. New York: Guilford Press
Golden NH, Ashtari M, Kohn MR, Patel M, Jacobson MS, Fletcher A, Shenker IR: Reversibility of cerebral ventricular enlargement in anorexia nervosa, demonstrated by quantitative magnetic resonance imaging. Journal of Pediatrics 1996; 128:296-301.
Golden NH: Osteopenia in adolescents with anorexia nervosa. Children's Hospital Quaterly 1992; 4:143-148.
Goldfarb, L. A., Fuhr, R., Tsujimoto, R. N., and Fischman, S. E. Systematic desensitization and relaxation as adjuncts in the treatment of anorexia nervosa: a preliminary study. Psychological Reports 1987; 60:511-518.
Goldstein, D. J., Wilson, M. G., Thomson, V. L., Potvin, J. H., and Rampey, A. H. Long-term fluoxetine treatment of bulimia nervosa. British Journal of Psychiatry 1995; 166:660-666.
Gordon CM, Grace E, Emans SJ, Feldman HA, Goodman E, Becker KA, Rosen CJ, Gundberg CM, LeBoff MS: Effects of oral dehydroepiandrosterone on bone density in young women with anorexia nervosa: a randomized trial. Journal of Clinical Endocrinology and Metabolism 2002; 87:4935-4941.
217 CONFIDENTIAL Second draft for consultation September 2003 Gordon, C. M., Grace, E., Emans, J., Goodman, E., Crawford, M. H., and Leboff, M. S. Changes in bone turnover markers and menstrual function after short-term oral DHEA in young women with anorexia nervosa. Journal of Bone and Mineral Research 1999; 14:136-145.
Gorin, Amy A. A controlled trial of cognitive-behavioral therapy with and without spousal involvement for binge eating disorder. Dissertation Abstracts International: Section B: the Sciences & Engineering 2001; 61.
Gowers, S., Norton, K., Halek, C., and Crisp, A. H. Outcome of outpatient psychotherapy in a random allocation treatment study of anorexia nervosa. International Journal of Eating Disorders 1994;15:165-177.
Gowers SG, Shore A. The stigma of eating disorders. Int J Clin Pract. 1999 Jul-Aug;53(5):386-8.
Gowers SG, Shore A. Development of weight and shape concerns in the aetiology of eating disorders.Br J Psychiatry. 2001 Sep;179:236-42. Review.
Griffiths, R. A. and Channon-Little, L. The hypnotizability of patients with bulimia nervosa and partial syndromes participating in a controlled treatment outcome study. Contemporary Hypnosis 1993;10:81-87.
Griffiths, R. A., Hadzi-Pavlovic, D., and Channon-Little, L. A controlled evaluation of hypnobehavioural treatment for bulimia nervosa: Immediate Pre-Post Treatment effects. European Eating Disorders Review 1994; 2:202-220.
Griffiths, R. A., Hadzi-Pavlovic, D., and Channon-Little, L. The short-term follow-up effect of hypnobehavioural and cognitive behavioural treatment for bulimia nervosa. European Eating Disorders Review 1996; 4:12-31.
Griffiths, R. A., Hazi-Pavlovic, D., and Channon-Little, L. Are there differences in response to psychological treatment for recruited and nonrecruited bulimic patients? European Eating Disorders Review 1997; 5:131-140.
Grinspoon S, Thomas E, Pitts S, Gross E, Mickley D, Miller K, Herzog D, Klibanski A: Prevalence and Predictive Factors for Regional Osteopenia in Women with Anorexia Nervosa. Annals of Internal Medicine 2000; 133:790-794.
Grinspoon, S., Baum, H., Lee, K., Anderson, E., Herzog, D., and Klibanski, A. Effects of short- term recombinant human insulin-like growth factor I administration on bone turnover in osteopenic women with anorexia nervosa. Journal of Clinical Endocrinology & Metabolism. 1996; 81:3864-3870.
Grinspoon S, Herzog D, Klibanski A Mechanisms and treatment options for bone loss in anorexia nervosa. Psychopharmacol Bull. 1997;33(3):399-404.Grinspoon, S.,
Thomas, L., Miller, K., Herzog, D., and Klibanski, A. Effects of recombinant human IGF-I and oral contraceptive administration on bone density in anorexia nervosa. Journal of Clinical Endocrinology & Metabolism. 2002; 87:2883-2891.
Gross, H., Ebert, M. H., and Faden, V. B. A double blind trial of Delta9-tetrahydrocannabinol in primary anorexia nervosa. Journal of Clinical Psychopharmacology 1983; 3:165-171.
Halbreich U, Kahn LS: Selective oestrogen receptor modulators--current and future brain and behaviour applications. Expert Opinion in Pharmacotherapy 2000; 1:1385-1398.
218 CONFIDENTIAL Second draft for consultation September 2003 Hall, A. and Crisp, A. H. Brief psychotherapy in the treatment of anorexia nervosa. Outcome at one year. British Journal of Psychiatry 1987; 151:185-191.
Hall A, Hay PJ. Eating disorder patient referrals from a population region 1977-1986. Psychol Med. 1991 Aug;21(3):697-701.
Halmi KA, Goldberg SC, Casper RC, Eckert ED, Davis JM. Pretreatment predictors of outcome in anorexia nervosa. British Journal of Psychiatry 1979; 134:71-78.
Halmi, K. A. and Goldberg, S. C. Cyproheptadine in anorexia nervosa. Psychopharmacology Bulletin 1978; 14:31-33.
Halmi, K. A., Eckert, E., and Falk, J. R. Cyproheptadine, an antidepressant and weight- inducing drug for anorexia nervosa. Psychopharmacology Bulletin. 1983; 19:103-105.
Halmi, K. A., Eckert, E., LaDue, T. J., and Cohen, J. Anorexia nervosa. Treatment efficacy of cyproheptadine and amitriptyline. Archives of General Psychiatry 1986; 43:177-181.
Hart, S., Field, T., Hernandez-Reif, M., Nearing, G., Shaw, S., Schanberg, S., and Kuhn, C. Anorexia nervosa symptoms are reduced by massage therapy. Eating Disorders 2001; 9:289- 299.
Harwood MP, Newton T. Dental aspects of bulimia nervosa: Implications for the health care team. European Eating Disorders Review 1995; 3(2):93-102.
Hay PJ. Eating disorders: Anorexia nervosa, bulimia nervosa and related syndromes - An overview of assessment and management. Australian Prescriber 1998; 21(4):100-103.
Hay PJ, Marley J, Lemar S. Covert eating disorders: The prevalence, characteristics and help- seeking of those with bulimic eating disorders in general practice. Primary Care Psychiatry 1998; 4(2):95-99.
(Hay & Bacaltchuk, 2002
Hay PJ, Bacaltchuk J. Psychotherapy for bulimia nervosa and binging. Cochrane Database Syst Rev. 2003;(1):CD000562.
Hebebrand J, Himmelmann GW, Herzog W, Herpertz-Dahlmann BM, Steinhausen HC, Amstein M, Seidel R, Deter HC, Remschmidt H, Schafer H. Prediction of low body weight at long-term follow-up in acute anorexia nervosa by low body weight at referral. American Journal of Psychiatry 1997; 154(4):566-569.
Hebebrand J, Himmelmann GW, Wewetzer C, Gutenbrunner C, Heseker H, Schafer H, Remschmidt H. Body weight in acute anorexia nervosa and at follow-up assessed with percentiles for the body mass index: implications of a low body weight at referral. International Journal of Eating Disorders. 1996; 19(4):347-357.
Hebebrand J, Wehmeier PM, Remschmidt H. Letter to the editor: Weight criteria for diagnosis of anorexia nervosa. American Journal of Psychiatry 2000; 157:1024.
Hebebrand, J., Himmelmann, G. W., Heseker, H., et al., Use of percentiles for the body mass index in anorexia nervosa: diagnostic, epidemiological, and therapeutic considerations. International Journal of Eating Disorders 1996; 19:359-369.
Herpertz-Dahlmann B, Muller B, Herpertz S, Heussen N, Hebebrand J, Remschmidt H. Prospective 10-year follow-up in adolescent anorexia nervosa--course, outcome, psychiatric
219 CONFIDENTIAL Second draft for consultation September 2003 comorbidity, and psychosocial adaptation. Journal of Child Psychology and Psychiatry. 2001; 42(5):603-612.
Herzog & Nussbaum, 1991
Herzog DB, Keller MB, Sacks NR, Yeh CJ, Lavori PW. 1992 Psychiatric comorbidity in treatment-seeking anorexics and bulimics. J Am Acad Child Adolesc Psychiatry. 1992 Sep;31(5):810-8.
Herzog DB, Nussbaum KM, Marmor AK. Comorbidity and outcome in eating disorders. Psychiatric Clinics of North America. 1996; 19(4):843-859.
Herzog W, Deter HC, Fiehn W, Petzold E. Medical findings and predictors of long-term physical outcome in anorexia nervosa: a prospective, 12-year follow-up study. Psychological Medicine 1997; 27(2):269-279.
Herzog W, Greenwood DN, Dorer DJ, et al: Mortality in Eating Disorders: A Descriptive Study. International Journal of Eating Disorders 2000; 28:20-26.
Herzog W, Schellberg D, Deter HC. First recovery in anorexia nervosa patients in the long- term course: a discrete-time survival analysis. Journal of Consulting and Clinical Psychology 1997; 65(1):169-177.
Herzog, T., Hartmann, A., and Falk, C. The short-term effects of psychodynamic inpatient treatment of anorexia nervosa with and without an explicit focus on eating pathology - A controlled study. Psychotherapie Psychosomatik Medizinische Psychologie 1996; 46:11-22.
Herzogs W, Deter HC, Fiehn W, Petzold E: Medical findings and predictors of long-term physical outcome in anorexia nervosa-a prospective, 12-year follow-up study. Psychological Medicine 2003; 27:269-279.
Hodes M, Timimi S, Robinson P: Children of mothers with eating disorders. A preliminary study. European Eating Disorders Review 1997; 5:1-24.
Hoek HW: The incidence and prevalence of anorexia nervosa in primary care. Psychological Medicine 1991; 21:455-460.
Hoek, H.W. Distribution of eating disorders. In: Maj, M. et al. (eds.) Eating disorders. 2003; John Wiley & Sons Ltd.
Hoek, H.W. The incidence and prevalence of anorexia nervosa in primary care. Psychological Medicine 1991; 21: 455-460.
Holderness CC, Brooks-Gunn J, Warren MP. Co-morbidity of eating disorders and substance abuse review of the literature. International Journal of Eating Disorders. 1994; 16(1):1-34.
Horne, R. L., Ferguson, J. M., Pope, H. G., Hudson, J. I., Lineberry, C. G., Ascher, J., and Cato, A. Treatment of bulimia with bupropion: a multicenter controlled trial. Journal of Clinical Psychiatry 1988; 49:262-266.
Howard WT, Evans KK, Quintero-Howard CV, Bowers WA, Andersen AE. Predictors of success or failure of transition to day hospital treatment for inpatients with anorexia nervosa. American Journal of Psychiatry 1999; 156(11):1697-1702.
Howlett M, McClelland L., Crisp AH: The cost of illness that defies. Postgraduate Medical Journal 1995; 71:36-39.
220 CONFIDENTIAL Second draft for consultation September 2003 Hsu LK, Crisp AH, Callender JS Psychiatric diagnoses in recovered and unrecovered anorectics 22 years after onset of illness: a pilot study. Compr Psychiatry. 1992 Mar- Apr;33(2):123-7.
Hsu, 1995
Hsu, L. K., Rand, W., Sullivan, S., Liu, D. W., Mulliken, B., McDonagh, B., and Kaye, W. H. Cognitive therapy, nutritional therapy and their combination in the treatment of bulimia nervosa. Psychological Medicine 2001;31:871-879.
Hsu, LKG Crisp AH & Harding B 1979 Outcome of anorexia nervosa Lancet i 61-65.
Hudson, J. I., McElroy, S., Raymond, N. C., Crow, S., Keck, P. E., Carter, M. D., Mitchell, J. E., Strakowski, M. D., Pope, H. G., Coleman, B. S., and Jonas, J. M. Fluvoxamine in the treatment of binge-eating disorder: a multicenter placebo-controlled, double-blind trial. American Journal of Psychiatry 1998; 155:1756-1762.
Huseman, C. A., Pearson, P. H., Madison, J., and Leuschen, M. P. Bulimia as a form of self- addiction: treatment with naltrexone hydrochloride (Trexan(TM)) - a pilot study. Clinical Trials Journal 1990; 27:77-83.
Iketani T, Kiriike N, Nakanishi S, Nakasuji T: Effects of weight gain and resumption of menses on reduced bone density in patients with anorexia nervosa. Biological Psychiatry 1995; 37:521-527.
Irwin M: Early Onset Anorexia Nervosa. Southern Medical Journal 1984; 77:611-614.
Isner J M, Roberts W C, Heymsfield S B, Yageer J: Anorexia Nervosa and Sudden Death. Annals of Internal Medicine 1985; 102:49-52.
Ivarsson T, Rastam M, Wentz E, Gillberg IC, Gillberg C. Depressive disorders in teenage- onset anorexia nervosa: a controlled longitudinal, partly community based study. Comprehensive Psychiatry 2000; 42(5):398-403
Jacobi, C., Dahme, D., and Dittmann, R. Cogntive-behavioural, fluoxetine and combined treatment for bulimia nervosa: short-and long-term results. European Eating Disorders Review 2002; 10:179-198.
Jansen, A., Elgersma, H., Nederkoorn, C., and Smeets, T. What makes treatment of bulimia nervosa successful? Paper presented at the European Association for Cognitive and Behavior Therapies. Maastricht, 18-21 2002.
Jansen A, Van den Hout MA, De Loof C, Zandbergen J, Griez E.A case of bulimia successfully treated by cue exposure. J Behav Ther Exp Psychiatry. 1989 Dec;20(4):327-32.Johnson WG, Jarrell MP, Chupurdia KM, Williamson DA.Repeated binge/purge cycles in bulimia nervosa: role of glucose and insulin. Int J Eat Disord. 1994 May;15(4):331-41.
Johnson JG, Spitzer RL, Williams JB.: Health problems, impairment and illnesses associated with bulimia nervosa and binge eating disorder among primary care and obstetric gynaecology patients. Psychological Medicine 2001; 31:1455-1466.
Jones, R., Peveler, R., Hope, R. A., and Fairburn, C. G. Changes during treatment for bulimia nervosa: A comparison of three psychological treatments. Behaviour Research and Therapy 1993; 1993:479-485.
221 CONFIDENTIAL Second draft for consultation September 2003 Kachele H, Kordy H, Mattias R, Research Group Treat. Therapy amount and outcome of inpatient psychodynamic treatment of eating disorders in Germany: data from a multi-center study. Psychotherapy Research 2001; 11(3) 239-257.
Kachele, H. Eine multizentrishe Studie zu Aufwand und Erfolg bei psychodynamischer Therapie von Essstorungen. Psychotherapy, Psychosomatik, Midizinische Psychologie 1999; 49:100-108.
Kanerva, R., Rissanen, A., and Sarna, S. Fluoxetine in the treatment of anxiety, depressive symptoms, and eating-related symptoms in bulimia nervosa. Journal of Psychiatry 1994; 49:237-242.
Kaplan AS. Psychological treatments for anorexia nervosa: A review of published studies and promising new directions 2002; 47: 235-242.
Kaplan DW, Blythe M, Diaz A, Feinstein RA, Fisher MM, Klein JD et al. Identifying and treating eating disorders. Pediatrics 2003; 111(1):204-211.
Katz, R. L., Keen, C. L., Litt, I. F., Hurley, L. S., Kellams-Harrison, K. M., and Glader, L. J. Zinc deficiency in anorexia nervosa. Journal of Adolescent Health Care 1987; 8:400-406.
Katzman DK, Lambe EK, Mikulis DJ, Ridgley JN, Goldbloom DS, Zipursky RB: Cerebral gray matter and white matter volume deficits in adolescent girls with anorexia nervosa. Journal of Pediatrics 1996; 129:794-803.
Kaye, W. H., Nagata, T., Weltzin, T. E., Hsu, L. K., Sokol, M. S., McConaha, C., Plotnicov, K. H., Weise, J., and Deep, D. Double-blind placebo-controlled administration of fluoxetine in restricting- and restricting-purging-type anorexia nervosa. Biological Psychiatry 2001; 49:644- 652.
Keel P K, Dorer D J, Eddy K T, et al: Predictors of Mortality in Eating Disorders. Archives of General Psychiatry 2003; 60:179-183.
Keel PK, Mitchell JE: Outcome in bulimia nervosa. American Journal of Psychiatry 1997; 154:313.
Keel PK, Mitchell JE, Miller KB, Davis TL, Crow SJ. Long-term outcome of bulimia nervosa. Arch Gen Psychiatry. 1999 Jan;56(1):63-9.
Keel, P. K., Mitchell, J. E., Davis, T. L., and Crow, S. J. Long-term impact of treatment in women diagnosed with bulimia nervosa. International Journal of Eating Disorders 2002; 31:151-158.
Keilen, M.; Treasure, T.; Schmidt, U.; Treasure, J. Quality of life measurements in eating disorders, angina, and transplant candidates: are they comparable? Journal of the Royal Society of Medicine, 1994; 87(8):441-444.
Kendall, T. J. G., Pilling, S., Barnes, T. R. E., et.al. (2003). Schizophrenia: Full national clinical guideline on core interventions in primary and secondary care. London: Gaskell.
King MB. Eating disorders in a general practice population. Prevalence, characteristics and follow up at 12 to 18 months. Psychological Medicine Monograph Suppl. 1989; 14:1-34.
Kingston K, Szmukler G, Andrewes D, Tress B, Desmond P: Neuropsychological and structural brain changes in anorexia nervosa before and after refeeding. Psychological Medicine 1996; 26: 15-28.
222 CONFIDENTIAL Second draft for consultation September 2003 Kirkley, B. G., Schneider, J. A., Agras, W. S., and Bachman, J. A. Comparison of two group treatments for bulimia. Journal of Consulting & Clinical Psychology 1985; 53:43-48.
Klibanski, A., Biller, B. M. K., Schoenfeld, D. A., Herzog, D. B., and Saxe, V. C. The effects of estrogen administration on trabecular bone loss in young women with anorexia nervosa. Journal of Clinical Endocrinology & Metabolism 1995; 80:898-904.
Kohn MR, Ashtari M, Golden NH, Schebendach J, Patel M, Jacobson MS, Shenker IR: Structural brain changes and malnutrition in anorexia nervosa. Annals of the New York Academy of Science. 1997; 817:398-399.
Koran, L.M.; Agras, W.S.; Rossiter, E.M.; et al. Comparing the cost-effectiveness of psychiatric treatments: bulimia nervosa. Psychiatry Research, 1995; 58: 13-21.
Krauth C, Buser K, Vogel H: How high are the costs of eating disorders-anorexia nervosa and bulimia nervosa-for German society? European Journal of Health Economics 2002; 3:244-250.
Kreipe RE, Birndorf SA. Eating disorders in adolescents and young adults. Medical Clinics of North America 2000; 84(4):1027-1049.
Krieg JC, Pirke KM, Lauer C, Backmund H: Endocrine, metabolic, and cranial computed tomographic findings in anorexia nervosa. Biological Psychiatry 1988; 23:377-387.
Kumetz, N. C. Bulimia: A descriptive and treatment study. PhD Thesis The University of Texas Health Science Center at Dallas. A descriptive and treatment study PhD Thesis The University of Texas Health Science Center at Dallas 1983.
Lacey JH. Bulimia nervosa, binge eating, and psychogenic vomiting: a controlled treatment study and long term outcome.
Br Med J (Clin Res Ed). 1983 May 21;286(6378):1609-13. Lacey, J. H. and Crisp, A. H. Hunger, food intake and weight: the impact of clomipramine on a refeeding anorexia nervosa population. Postgraduate Medical Journal 1980; 56:79-85.
Lacey JH, Smith G.Bulimia nervosa. The impact of pregnancy on mother and baby. Br J Psychiatry. 1987 Jun;150:777-81.
Lacey 1997
Laederach-Hofmann, K., Graf, C., Horber, F., Lippuner, K., Lederer, S., Michel, R., and Schneider, M. Imipramine and diet counseling with psychological support in the treatment of obese binge eaters: a randomized, placebo-controlled double-blind study. International Journal of Eating Disorders 1999; 26:231-244.
Laessle, R. G., Beumont, P. J. V., Butow, P., Lennerts, W., O'Conner, M., Pirke, K. M., Touyz, S. W., and Waadt, S. A comparison of nutritional management with stress management in the treatment of bulimia nervosa. British Journal of Psychiatry 1991; 159:250-261.
Laessle, R. G., Waadt, S., and Pirke, K. M. A structured behaviorally orientated group treatment for bulimia nervosa. Psychotherapy & Psychosomatics 1987; 48:141-145.
Lambe EK, Katzman DK, Mikulis DJ, Kennedy SH, Zipursky RB: Cerebral gray matter volume deficits after weight recovery from anorexia nervosa. Archives of General Psychiatry 1997; 54:537-542.
223 CONFIDENTIAL Second draft for consultation September 2003 Lask, B., Fosson, A., Rolfe, U., and Thomas, S. Zinc deficiency and childhood-onset anorexia nervosa. Journal of Clinical Psychiatry. 1993; 54:63-66.
Le Grange, D., Eisler, I., Dare, C., and Russell, G. Evaluation of family treatments in adolescent anorexia nervosa: a pilot study. International Journal of Eating Disorders 1992; 12:347-357.
Le Grange, Daniel, Gelman, Tamara. “PATIENTS' PERSPECTIVE OF TREATMENT IN EATING DISORDERS: A PRELIMINARY STUDY”, South African Journal of Psychology, 00812463, Sep98, Vol. 28, Issue 3.
Lee, N. F. and Rush, A. J. Cognitive-behavioural group therapy for bulimia. International Journal of Eating Disorders 1986; 5:599-615.
Leibowitz SF, Alexander JT: Hypothalamic serotonin in control of eating behavior, meal size, and body weight. Biological Psychiatry 1998; 44:851-864.
Leitenberg, H., Rosen, J. C., Gross, J., Nudelman, S., and Vara, L. S. Exposure plus response- prevention treatment of bulimia nervosa. Journal of Consulting & Clinical Psychology 1988; 56:535-541.
Leitenberg, H., Rosen, J. C., Wolf, J., Vara, L. S., Detzer, M. J., and Srebnik, D. Comparison of cognitive-behavior therapy and desipramine in the treatment of bulimia nervosa. Behavioral Research and Therapy 1994; 32:37-45.
Lemouchoux, C.; Millar, H.; Naji, S. Eating disorders in Scotland: starved of resources? Psychiatric Bulletin, 2001; 21(7):256-260
Lock, J; Le Grange, D. Can family-based treatment of anorexia be manualized? Journal of Psychotherapy Practice Research 2001 Fall; 10(4)253-61.
Loeb, K. L., Wilson, G. T., Gilbert, J. S., and Labouvie, E. Guided and unguided self-help for binge eating. Behaviour Research and Therapy 2000; 38:259-272.
Lowe B, Zipfel S, Buchholz C, Dupont Y, Reas DL, Herzog W. Long-term outcome of anorexia nervosa in a prospective 21-year follow-up study. Psychological Medicine. 2001; 31(5):881-90.
Lucas AR, Melton LJ, III, Crowson CS, O'Fallon WM: Long-term fracture risk among women with anorexia nervosa: a population-based cohort study. ., 74, 972-977. Mayo Clin.Proc 1999; 74:972-977.
Luck AJ, Morgan JF, Reid F, O'Brien A, Brunton J, Price C, Perry L, Lacey JH. The SCOFF questionnaire and clinical interview for eating disorders in general practice: comparative study. BMJ. 2002 Oct 5;325(7367):755-6
Mann, T. Clinical guidelines: using clinical guidelines to improve patient care within the NHS. Department of Heath NHS Executive, 1996.
Maugars YM, Berthelot JM, Forestier R, Mammar N, Lalande S, Venisse JL, Prost AM: Follow- up of bone mineral density in 27 cases of anorexia nervosa. European Journal of Endocrinology 1996; 135:591-597.
MacDonald, C. Treatment resistance in anorexia nervosa and the pervasiveness of ethics in clinical decision-making. Canadian Journal of Psychiatry 2002:47:267-270.
224 CONFIDENTIAL Second draft for consultation September 2003 McCann, U. D. and Agras, W. S. Successful treatment of nonpurging bulimia nervosa with desipramine: a double-blind, placebo-controlled study. American Journal of Psychiatry 1990;147:1509-1513.
McClain, C. J., Stuart, M. A., Vivian, B., McClain, M., Talwalker, R., Snelling, L., and Humphries, L. Zinc status before and after zinc supplementation of eating disorder patients. Journal of the American College of Nutrition 1992; 11, 694-700.
McElroy, S. L., Casuto, L. S., Nelson, E. B., Lake, K. A., Soutullo, C. A., Keck, P. E., Jr., and Hudson, J. I. Placebo-controlled trial of sertraline in the treatment of binge eating disorder. American Journal of Psychiatry 2000; 157:1004-1006.
McElroy, S., Arnold, L. M., Shapira, N. A., Keck, P. E., Rosenthal, N. R., Karim, R., Kamin, M., and Hudson, J. I. Topiramate in the treatment of binge eating disorder associated with obesity: a randomised, placebo-controlled trial. American Journal of Psychiatry; In Press.
McIntosh, V. V., Jordan, J., Carter, F. A., Luty, S. E., McKenzie, J. M., Bulik, C. M., Frampton, C. M. A., and Joyce, P. R. Three psychotherapies for anorexia nervosa: a randomized controlled trial. Submitted for publication.
Meads, C.; Gold, L.; Burls, A. How effective is outpatient care compared to inpatient care for the treatment of anorexia nervosa? A systematic review. European Eating Disorders Review, 2001; 9: 229-241.
Mehler P S. Diagnosis and Care of Patients with Anorexia Nervosa in Primary Care Settings. Annals of Internal Medicine 2001; 1048-1059.
Mehler, P. S. & Linas, S. Use of a proton-pump inhibitor for metabolic disturbances associated with anorexia nervosa. New England Journal of Medicine 2002; 347: 373-374.
Mitchell JE, Groat R.A placebo-controlled, double-blind trial of amitriptyline in bulimia. J Clin Psychopharmacol. 1984 Aug;4(4):186-93.
Mitchell 1998
Mitchell JE, Hatsukami D, Eckert ED, Pyle RL.Characteristics of 275 patients with bulimia.Am J Psychiatry. 1985 Apr;142(4):482-5.
Mitchell JE, Pomeroy C, Seppala M, Huber M.Diuretic use as a marker for eating problems and affective disorders among women. J Clin Psychiatry. 1988 Jul;49(7):267-70.
Mitchell 1995
Mitchell, J. E., Christenson, G., Jennings, J., Huber, M., Thomas, B., Pomeroy, C., and Morley, J. A placebo-controlled, double-blind crossover study of naltrexone hydrochloride in outpatients with normal weight bulimia. Journal of Clinical Psychopharmacology 1989; 9:94- 97.
Mitchell, J. E., Fletcher, L., Hanson, K., Mussell, M. P., Seim, H., Crosby, R., and Al Banna, M. The relative efficacy of fluoxetine and manual-based self-help in the treatment of outpatients with bulimia nervosa. Journal of Clinical Psychopharmacology 2001; 21:298-304.
Mitchell, J. E., Pyle, R. L., Eckert, E. D., Hatsukami, D., Pomeroy, C., and Zimmerman, R. A comparison study of antidepressants and structured intensive group psychotherapy in the treatment of bulimia nervosa. Archives of General Psychiatry 1990; 47:149-157.
225 CONFIDENTIAL Second draft for consultation September 2003 Mitchell J.E., Pomeroy C., & Adson D. (1997). Managing medical complications in Garner D. & Garfinkel P. (Eds) Handbook of treatment for eating disorder, 2nd Edition. New York, Guilford Press. Pp. 383-393.
Mitchell, J.E.; Peterson, C.B.; Agras, S. Cost-effectiveness of psychotherapy for eating disorders. In: Miller, N.E. (ed.) Cost-effectiveness of psychotherapy, 1999; Oxford University Press, Oxford
Mitchell JE, Halmi K, Wilson GT, Agras WS, Kraemer H, Crow S. A randomized secondary treatment study of women with bulimia nervosa who fail to respond to CBT. Int J Eat Disord. 2002 Nov;32(3):271-81.
Morgan HG, Purgold J, Welbourne J. Management and outcome in anorexia nervosa. A standardized prognostic study. British Journal of Psychiatry. 1983; 143:282-7.
Morgan HG, Russell GF. Value of family background and clinical features as predictors of long-term outcome in anorexia nervosa: four-year follow-up study of 41 patients. Psychological Medicine. 1975; 5(4):355-71.
Morgan JF, Lacey JH, Sedgwick PM: Impact of pregnancy on bulimia nervosa. British Journal of Psychiatry 1999; 174:135-140.
Morgan JF, Reid F and Lacey JH. The SCOFF questionnaire: assessment of a new screening tool for eating disorders. British Medical Journal 1999; 319:1467-1468.
Mumford, DB and Whitehouse, AM. Increased prevalence of bulimia nervosa among Asian schoolgirls. British Medical Journal. 1998; 297, 718-719.
Nauta, H., Hospers, H., Kok, G., and Jansen, A. A comparison between a cognitive and a behavioral treatment for obese binge eaters and obese non-binge eaters. Behavior Therapy 2000;31:441-461.
Netten, A.; Rees, T.; Harrison, G. Unit costs of health and social care. 2002; Personal Social Services Research Unit, Canterbury.
Neumärker K J: Mortality and Sudden Death in Anorexia Nervosa. International Journal of Eating Disorders 1997; 21:205-212.
Neumärker K J: Mortality Rates and Causes of Death. European Eating Disorders Review 2000; 8:181-187.
Neumarker KJ, Dudeck U, Meyer U, Neumarker U, Schulz E, Schonheit B: Anorexia nervosa and sudden death in childhood: clinical data and results obtained from quantitative neurohistological investigations of cortical neurons. Eur.Arch.Psychiatry Clin.Neurosci. 1997; 247:16-22.
Nicholls D, Stanhope R: Medical Complications of Anorexia Nervosa in Children and Young Adolescents. European Eating Disorders Review 2000; 8:170-178.
Nielsen S, Emborg C, Molbak AG: Mortality in concurrent type 1 diabetes and anorexia nervosa. Diabetes Care 2002; 25:309-312.
Nielsen S, Moller-Madsen S, Isager T, Jorgensen J, Pagsberg K, Theander S: Standardized mortality in eating disorders--a quantitative summary of previously published and new evidence. Journal of Psychosomatic Research 1998; 44: 413-34
226 CONFIDENTIAL Second draft for consultation September 2003 Nielsen S: Eating disorders in females with type 1 diabetes: An update of a meta-analysis. European Eating Disorders Review 2002; 10:241-254.
Nielsen S: Epidemiology and Mortality of Eating Disorders. Eating Disorders 2001; 24:201- 214.
Noordenbos G. Eating disorders in primary care: Early identification and intervention by general practitioners. In: Vandereycken W, Nordenbos G, eds. The prevention of eating disorders. London: Athlone Press, 1988; 214–229. Nussbaum M, Baird D, Sonnenblick M, Cowan K, Shenker IR. Short stature in anorexia nervosa patients. Journal of Adolescent Health Care 1985;6(6):453-455.
Office of Health Economics. . Eating disorders. 111. 1994. London, OHE.
Ogg EC, Millar HR, Pusztai EE, Thom AS. General practice consultation patterns preceding diagnosis of eating disorders. International Journal of Eating Disorders 1997; 22(1):89-93.
O'Herlihy, A, Worrall, A., Banerjee, S., and et al. National In-patient Child and Adolescent Psychiatry Study (NICAPS): Final Report to the Department of Health. London. Royal College of Psychiatrists' Research Unit. 2001.
Ostuzzi R, Didonna F, Micciolo R. One-year weight follow-up in anorexia nervosa after inpatient psycho-nutritional rehabilitative treatment. Eat Weight Disord. 1999; 4(4):194-7.
Patton GC. Mortality in eating disorders. Psychological Medicine. 1988; 18(4):947-51.
Perry L, Morgan J, Reid F, Brunton J, O'Brien A, Luck A, Lacey H. Screening for symptoms of eating disorders: reliability of the SCOFF screening tool with written compared to oral delivery. Int J Eat Disord. 2002 Dec;32(4):466-72.
Pertschuk, M. J., Forster, J., Buzby, G., and Mullen, J. L. The treatment of anorexia nervosa with total parenteral nutrition. Biological Psychiatry. 1981; 16, 539-550.
Pike, K. M., Walsh, B. T., Vitousek, K., Wilson, G. T., and Bauer, J. Cognitive Behavioral Therapy in the Post-Hospital Treatment of Anorexia Nervosa. Submitted for publication.
Pike, KM, Walsh, BT, Vitousek, K, Wilson, GT, and Bauer, J. Cognitive behavioral therapy in the post-hospital treatment of anorexia nervosa. American Journal of Psychiatry . In Press.
Pillay, M. and Crisp, A. H. The impact of social skills training within an established in-patient treatment programme for anorexia nervosa. British Journal of Psychiatry 1981;139:533-539.
Pope & Hudson, 1982
Pope, H. G. Jr, Keck, P. E., McElroy, S., and Hudson, J. I. A placebo-controlled study of trazodone in bulimia nervosa. Journal of Clinical Psychopharmacology 1989; 9:254-259.
Pope, H. G., Hudson, J. I., Jonas, J. M., and Yurgelum-Todd, D. Antidepressant treatment of bulimia: A two-year follow-up study. Journal of Clinical Psychopharmacology 1985; 5:320- 327.
Pope, H. G., Hudson, J. I., Jonas, J. M., and Yurgelum-Todd, D. Bulimia treated with imipramine: a placebo-controlled, double-blind study. American Journal of Psychiatry 1983; 140:554-558.
227 CONFIDENTIAL Second draft for consultation September 2003 Pyle, R. L., Mitchell, J. E., Eckert, E. D., Hatsukami, D., Pomeroy, C., and Zimmerman, R. Mainenance treatment and 6-month outcome for bulimic patients who respond to initial treatment. American Journal of Psychiatry 1990;147:871-875.
Ratnasuriya RH, Eisler I, Szmukler GI, Russell GF. Anorexia nervosa: outcome and prognostic factors after 20 years. British Journal of Psychiatry. 1991; 158:495-502.
Reeves, R. S., McPherson, R. S., Nichaman, M. Z., Harrist, R. B., Foreyt, J. P., and Goodrick, G. K. Nutrient intake of obese female binge eaters. Journal of the American Dietetic Association 2001; 101:209-215.
Remschmidt H, Wienand F, Wewetzer C. Long term prognosis in anorexia nervosa. Monatsschr Kinderheilkd. 1988 Nov;136(11): 726-31. Review. German.
Ricca V, Mannucci E, Mezzani B, et al: Psychopathological and clinical features of outpatients with an eating disorder not otherwise specified. Eat Weight Disord 2001; 6:157-165.
Ricca V, Mannucci E, Moretti S, Di Bernardo M, Zucchi T, Cabras PL, and Rotella CM. Screening for binge eating disorder in obese outpatients. Comprehensive Psychiatry 2000; 41: 111-115 .
Rigotti NA, Neer RM, Skates SJ, Herzog DB, Nussbaum SR: The clinical course of osteoporosis in anorexia nervosa. A longitudinal study of cortical bone mass. Journal of the American Medical Association 1991; 265:1133-1138.
Robb, A. S., Silber, T. J., Orrell-Valente, J. K., Valadez-Meltzer, A., Ellis, N., Dadson, M. J., and Chatoor, I. Supplemental nocturnal nasogastric refeeding for better short-term outcome in hospitalized adolescent girls with anorexia nervosa. American Journal of Psychiatry 2002; 159:1347-1353.
Robin, A. L., Siegel, P. T., Koepke, T., Moye, A. W., and Tice, S. Family therapy versus individual therapy for adolescent females with anorexia nervosa. Journal of developmental and behavioral pediatrics 1994; 15:111-116.
Robin, A. L., Siegel, P. T., Moye, A. W., Gilroy, M., Dennis, A. B., and Sikand, A. A controlled comparison of family versus individual therapy for adolescents with anorexia nervosa. American Academy of Child and Adolescent Psychiatry 1999; 38:1482-1489.
Robinson P.: Day Treatments, Handbook of Eating Disorders, 2nd Edition. Edited by Treasure J SU&vFE. Wiley, 2003.
Robinson PH. Review article: recognition and treatment of eating disorders in primary and secondary care. Alimentary Pharmacology & Therapeutics 2000; 14(4):367-377.
Romano, S. J., Halmi, K. A., Sarkar, N. P., Koke, S. C., and Lee, J. S. A placebo-controlled study of fluoxetine in continued treatment of bulimia nervosa after successful acute fluoxetine treatment. American Journal of Psychiatry 2002; 159:96-102.
Rome ES. Eating disorders in adolescents and young adults: what's a primary care clinician to do? Clevland Clinic Journal of Medicine 1996; 63(7):387-395.
Rosenvinge, 2001
Rosenvinge & Klusmeier, 2000
228 CONFIDENTIAL Second draft for consultation September 2003 Rosenvinge JH, Martinussen M, Ostensen E. The comorbidity of eating disorders and personality disorders: a meta-analytic review of studies published between 1983 and 1998. Eat Weight Disord. 2000 Jun;5(2):52-61.
Ruggiero, G. M., Laini, V., Mauri, M. C., Ferrari, V. M., Clemente, A., Lugo, F., Mantero, M., Redaelli, G., Zappulli, D., and Cavagnini, F. A single blind comparison of amisulpride, fluoxetine and clomipramine in the treatment of restricting anorectics. Progress in Neuro- Psychopharmacology & Biological Psychiatry 2001; 25:1049-1059.
Russell, 1979
Russell G F M: Pre-Menarchal Anorexia Nervosa and Its Sequelae. Journal of Psychiatric Research 1985; 19:363-369.
Russell, 2001
Russell GF, Treasure J, Eisler I: Mothers with anorexia nervosa who underfeed their children: their recognition and management. Psychological Medicine 1998; 28:93-108.
Russell J, Gross G. Anorexia nervosa and body mass index. American Journal of Psychiatry. 2000; 157(12):2060.
Russell, G. F., Szmukler, G. I., Dare, C., and Eisler, I. An evaluation of family therapy in anorexia nervosa and bulimia nervosa. Archives of General Psychiatry 1987;44:1047-1056.
Sabine, E. J., Yonace, A., Farrington, A. J., Barrnat, K. H., and Wakeling, A. Bulimia nervosa: a placebo-controlled double-blind therapeutic trial of mianserin. British Journal of Clinical Pharmacology 1983; 15:195S-202S.
Saccomani L, Savoini M, Cirrincione M, Vercellino F, Ravera G. Long-term outcome of children and adolescents with anorexia nervosa: study of comorbidity. Journal of Psychosomatic Research. 1998; 44(5):565-571.
Safer, D. L., Telch, C. F., and Agras, W. S. Dialectical behavior therapy for bulimia nervosa. American Journal of Psychiatry 2001; 158:632-634.
Santonastaso P, Favaretto G, Canton G. Anorexia nervosa in Italy: clinical features and outcome in a long-term follow-up study. Psychopathology. 1987; 20(1):8-17.
Santonastaso P, Pantano M, Panarotto L, Silvestri A. A follow-up study on anorexia nervosa: clinical features and diagnostic outcome. European Psychiatry. 1991; 6:177-185.
Schork EJ, Eckert ED, Halmi KA. The relationship between psychopathology, eating disorder diagnosis, and clinical outcome at 10-year follow-up in anorexia nervosa. Comprehensive Psychiatry. 1994; 35(2):113-123.
Seeman E, Szmukler GI, Formica C, Tsalamandris C, Mestrovic R: Osteoporosis in anorexia nervosa: the influence of peak bone density, bone loss, oral contraceptive use, and exercise. Journal of Bone Mineral Research 1992; 7:1467-1474.
Serfaty, M. A., Turkington, D., Heap, M., Ledsham, L., and Jolley, E. Cognitive therapy versus dietary counselling in the outpatient treatment of anorexia nervosa: effects of the treatment phase. European Eating Disorders Review 1999; 7:334-350.
Serfaty et al., 2002
229 CONFIDENTIAL Second draft for consultation September 2003 Silber, Tomas J; D'Angelo, Lawrence J. The role of the primary care physician in the diagnosis and management of anorexia nervosa. Psychosomatics; 1991; 32(2): 221-225.
Solanto, M. V., Jacobson, M. S., Hellers, L., Golden, N. H., and Hertz, S. Rate of weight gain of inpatients with anorexia nervosa under behavioral contracts. Pediatrics 1994; 93:989-991.
Soyka LA, Grinspoon S, Levitsky LL, Herzog DB, Klibanski A: The effects of anorexia nervosa on bone metabolism in female adolescents. Journal of Clinical Endocrinology and Metabolism 1999; 84:4489-4496.
Sriegel-Moore R., Silberstein L.R., Rodin J.: and (1986). Toward an understanding of risk factors for bulimia. American Psychologist 1986; 41:246-263.
Stefanis N, Mackintosh C, Abraha HD, Treasure J, Moniz C: Dissociation of bone turnover in anorexia nervosa. Annals of Clinical Biochemistry 1998; 35:709-716.
Steiner H, Smith C, Rosenkranz RT, Litt I: The early care and feeding of anorexics. Child Psychiatry and Human Development 1991; 21:163-167.
Steinhausen H C: The Outcome of Anorexia Nervosa in The 20th Century. American Journal of Psychiatry 2002; 159:1284-1293.
Steinhausen, 1995).
Steinhausen HC, Glanville K. A long-term follow-up of adolescent anorexia nervosa. Acta Psychiatrica Scandinavica. 1983; 68(1):1-10.
Steinhausen HC, Seidel R. Outcome in adolescent eating disorders. International Journal of Eating Disorders 1993; 14(4):487-96.
Stice E: Risk and maintenance factors for eating pathology: a meta-analytic review. Psychological Bulletin 2002; 128:825-848.
Stice, E., & Agras, W.S. Predicting onset and cessation of bulimic behaviors during adolescence: A longitudinal grouping analysis. Behavior Therapy, 29, 257-276, 1998.
Stonehill E, Crisp AH. Psychoneurotic characteristics of patients with anorexia nervosa before and after treatment and at follow-up 4-7 years later. Journal of Psychosomatic Research. 1977; 21(3):189-93.
Striegel-Moore RH, Leslie D, Petrill SA, Garvin V, Rosenheck RA: One-year use and cost of inpatient and outpatient services among female and male patients with an eating disorder: Evidence from a national database of health insurance claims. International Journal of Eating Disorders 2000; 23:161-167.
Striegel-Moore RH, Silberstein LR, Rodin J. 1986 Toward an understanding of risk factors for bulimia. Am Psychol. 1986 Mar;41(3):246-63.
Strober M, Freeman R, Morrel W: The long-term course of severe anorexia nervosa in adolescents: survival analysis of recovery, relapse and outcome predictors over 10-15 years in a prospective study. International Journal of Eating Disorders 1997; 22:339.
Strober M, Freeman R, Morrell W. Atypical anorexia nervosa: separation from typical cases in course and outcome in a long-term prospective study. International Journal of Eating Disorders. 1999; 25(2):135-42.
230 CONFIDENTIAL Second draft for consultation September 2003 Strober M, Freeman R, Morrell W. The long-term course of severe anorexia nervosa in adolescents: survival analysis of recovery, relapse, and outcome predictors over 10-15 years in a prospective study. International Journal of Eating Disorders. 1997; 22(4):339-60.
Strober M.: Stressful life events associated with bulimia in anorexia nervosa Empirical findings and theoretical speculations. International Journal of Eating Disorders 1984; 3:3-16.
Sullivan P F: Mortality in Anorexia Nervosa. American Journal of Psychiatry 1995; 152:1073- 1074.
Sundgot-Borgen, J., Rosenvinge, J. H., Bahr, R., and Schneider, L. S. The effect of exercise, cognitive therapy, and nutritional counseling in treating bulimia nervosa. Medicine & Science in Sports & Exercise 2002; 34:190-195.
Swayze VW, Andersen A, Arndt S, Rajarethinam R, Fleming F, Sato Y, Andreasen NC: Reversibility of brain tissue loss in anorexia nervosa assessed with a computerized Talairach 3-D proportional grid. Psychological Medicine 1996; 26:381-390.
Tanaka H, Kiriike N, Nagata T, Riku K. Outcome of severe anorexia nervosa patients receiving inpatient treatment in Japan: an 8-year follow-up study. Psychiatry and Clinical Neurosciences. 2001; 55(4):389-396.
Tanaka et al., 2002
Tchanturia K, Morris R, Surguladze S, and Treasure J. Perceptual and cognitive set shifting tasks in acute anorexia nervosa and following recovery. Eat.Weight.Disord. In Press.
Tchanturia K, Serpell L, Troop N, Treasure J: Perceptual illusions in eating disorders: rigid and fluctuating styles. J Behav.Ther Exp.Psychiatry 2001; 32:107-115.
Telch, C. F., Agras, W. S., and Linehan, M. M. Dialectical behavior therapy for binge eating disorder. Journal of Consulting & Clinical Psychology 2001; 69:1061-1065.
Telch, C. R., Agras, W. S., Wilfley, D. E., and Kenardy, J. Group cognitive behavioural treatment for the non-purging bulimic: an initial evaluation. Journal of Consulting & Clinical Psychology 1990; 58:629-635.
Theander S. Anorexia nervosa. A psychiatric investigation of 94 female patients. Acta Psychiatrica Scandinavica Suppl. 1970; 214:1-194.
Thien, V., Thomas, A., Markin, D., and Birmingham, C. L. Pilot study of a graded exercise program for the treatment of anorexia nervosa. International Journal of Eating Disorders 2000; 28:101-106.
Toner BB, Garfinkel PE, Garner DM. Long-term follow-up of anorexia nervosa. Psychosomatic Medicine. 1986; 48(7):520-9
Tozzi F, Sullivan P, Fear J., McKenzie J, Bulik C.M.: Causes and recovery in anorexia nervosa: the patient's perspective. International Journal of Eating Disorders 2003; 33:143-154.
Treasure J, Gavan K, Todd G, Schmidt U Changing the environment in eating disorders\; working with carers/families to improve motivation and facilitate change European Eating Disorders Review 2003; 11, 25-37
231 CONFIDENTIAL Second draft for consultation September 2003 Treasure J, Murphy T, Szmukler G, Todd G, Gavan K, Joyce J: The experience of caregiving for severe mental illness: a comparison between anorexia nervosa and psychosis. Social Psychiatry and Psychiatric Epidemiology 2001; 36:343-347.
Treasure JL, Russell GF, Fogelman I, Murby B: Reversible bone loss in anorexia nervosa. British Medical Journal (Clin.Res.Ed) 1987; 295:474-475.
Treasure JL, Russell GF: Intrauterine growth and neonatal weight gain in babies of women with anorexia nervosa. British Medical Journal (Clin.Res.Ed) 1988; 296:1038.
Treasure JL: Anorexia nervosa and bulimia, 1st Edition. Edited by Royal College Psychiatrists. London, Gaskell Press, 1998, pp 858-902.Treasure & Schmidt, 2002 Anorexia nervosa. Clin Evid. 2002 Dec;(8):903-13.
Treasure & Schmidt, 2001 TREASURE, J. & SCHMIDT, U. (2001). Needs assessment and eating disorders. In: Measuring Mental Health Needs (eds G. Thornicroft, C. R. Brewin, & J. Wing) (2nd edn) London: Gaskell (in press).
Treasure, J. L. & Ward, A. (1997) Cognitive analytical therapy (CAT) in eating disorders. Clinical Psychology and Psychotherapy, 4, 62-71
Treasure & Schmidt, 2003
Treasure & Bauer, 2003
Treasure, J., Katzman, M., Schmidt, U., Troop, N., Todd, G., and de Silva, P. Engagement and outcome in the treatment of bulimia nervosa: first phase of a sequential design comparing motivation enhancement therapy and cognitive behavioural therapy. Behaviour Research and Therapy 1999;37:405-418.
Treasure, J., Schmidt, U., Troop, N., Tiller, J., Todd, G., and Turnbull, S. Sequential treatment for bulimia nervosa incorporating a self-care manual. British Journal of Psychiatry 1996; 168:84-98.
Treasure, J., Schmidt, U., Troop, N., Tiller, J., Todd, G., Keilen, M., and Dodge, E. First step in managing bulimia nervosa: controlled trial of a therapeutic manual. British Medical Journal 1994; 308:686-689.
Treasure, J., Todd, G., Brolly, M., Tiller, J., Nehmed, A., and Denman, F. A pilot study of a randomised trial of cognitive analytical therapy vs educational behavioral therapy for adult anorexia nervosa. Behavioral Research and Therapy 1995; 33:363-367.
Troop, N., Schmidt, U., Tiller, J., Todd, G., Keilen, M., and Treasure, J. Compliance with a self- care manual for bulimia nervosa: Predictors and outcome. British Journal of Clinical Psychology 1996; 35:435-438.
Turnball, S. J., Schmidt, U., Troop, N., Tiller, J., Todd, G., and Treasure, J. Predictors of outcome for two treatments for bulimia nervosa. International Journal of Eating Disorders 1997; 21:17-22.
Turnbull S, Ward A, Treasure J, Jick H, Derby L: The demand for eating disorder care. British Journal of Psychiatry 1996; 169:705-712.
Turnbull W, Treasure J, Derby: The demand for eating disorders care. Epidemiological study using the General Practice Research Database. British Journal of Psychiatry 1996; 169:705-712.
232 CONFIDENTIAL Second draft for consultation September 2003 Turner H and Bryant-Waugh R. Eating disorder not otherwise specified (EDNOS) profiles of clients presenting at a community eating disorder service. European Eating Disorders Review . In Press.
Valla A, Groenning IL, Syversen U, Hoeiseth A: Anorexia nervosa: slow regain of bone mass. Osteoporosis International 2000; 11:141-145.
Vandereycken, W. and Pierloot, R. Pimozide combined with behavior therapy in the short- term treatment of anorexia nervosa. A double-blind placebo-controlled cross-over study. Acta Psychiatrica Scandinavica 1982; 66:445-450.
Vandereycken, W. Neuroleptics in the short-term treatment of anorexia nervosa. A double- blind placebo-controlled study with sulpiride. British Journal of Psychiatry 1984; 144:288-292.
Vandereycken & Pierloot, 1993
Vestergaard P, Emborg C, Stoving RK, Hagen C, Mosekilde L, Brixen K: Fractures in patients with anorexia nervosa, bulimia nervosa, and other eating disorders--a nationwide register study. International Journal Eating Disorders 2002; 32:301-308.
Vogelantz-Holm N., Wonderlich S.A, Lewis B, et al: Longitudinal predictors of binge eating, intense dieting, and weight concerns in a national sample of women. Behaviour Therapy 2000; 31:221-235.
Vos T, et al: The burden of mental disorders in Victoria. Social Psychiatry Psychiatric Epidemiology 2001; 36:53-62.
Waaddegaard et al., 1999
Wallin, U., Kronovall, P., and and, Majewski M. L. Body awareness therapy in teenage anorexia nervosa: outcome after 2 years. European Eating Disorders Review 2000; 8:19-30.
Walsh JM, Wheat ME, Freund K. Detection, evaluation, and treatment of eating disorders the role of the primary care physician. Journal of General Internal Medicine 2000; 15(8):577-590.
Walsh, B. T. Medication in the treatment of bulimia. Adolescent Psychiatry 1986; 13:437-445.
Walsh, B. T., Fairburn, C. G., Mickley, D., Sysko, R., and Parides, M. K. Treatment of bulimia nervosa in a primary care setting. Submitted.
Walsh, B. T., Gladis, M., Roose, S. P., Stewart, J. W., and Glassman, A. H. A controlled trial of phenelzine in bulimia. Psychopharmacology Bulletin. 1987;23:49-51.
Walsh, B. T., Gladis, M., Roose, S. P., Stewart, J. W., and Glassman, A. H. Phenelzine vs placebo in 50 patients with bulimia. Archives of General Psychiatry 1988; 45:471-475.
Walsh, B. T., Hadigan, C. M., Devlin, M. J., Gladis, M., and Roose, S. P. Long-term outcome of antidepressant treatment for bulimia nervosa. Journal of Psychiatry 1991; 148:1206-1212.
Walsh, B. T., Stewart, J. W., Roose, S. P., Gladis, M., and Glassman, A. H. Treatment of bulimia with phenelzine. A double-blind, placebo-controlled study. Archives of General Psychiatry. 1984; 41:1105-1109.
Walsh, B. T., Stewart, J. W., Roose, S. P., Gladis, M., and Glassman, A. H. A double-blind trial of phenelzine in bulimia. Journal of Psychiatric Research. 1985; 19:485-489.
233 CONFIDENTIAL Second draft for consultation September 2003 Walsh, B. T., Wilson, G. T., Loeb, K. L., Devlin, M. J., Pike, K. M., Roose, S. P., Fleiss, J., and Waternaux, C. Medication and psychotherapy in the treatment of bulimia nervosa. American Journal of Psychiatry 1997; 154:523-531.
Walsh BT, Stewart JW, Wright L, Harrison W, Roose SP, Glassman AH.1982 Treatment of bulimia with monoamine oxidase inhibitors.Am J Psychiatry. 1982 Dec;139(12):1629-30.
Ward et al., 1996
Ward A, Brown N, Treasure J: Persistent osteopenia after recovery from anorexia nervosa. International Journal of Eating Disorders 1997; 22:71-75.
Weizman, A., Tyano, S., Wijsenbeek, H., and Ben David, M. Behavior therapy, pimozide treatment and prolactin secretion in anorexia nervosa. Psychotherapy & Psychosomatics 1985; 43:136-140.
Wezel-Meijler G, Wit JM: The offspring of mothers with anorexia nervosa: a high-risk group for undernutrition and stunting? European Journal of Peadiatrics 1989; 149:130-135.
Wilfley, D. E., Agras, W. S., Telch, C. F., Rossiter, E. M., Schneider, J. A., Cole, A. G., Sifford, L. A., and Raeburn, S. D. Group cognitive behavioural and group interpersonal psychotherapy for thenonpurging bulimic individual: a controlled comparison. Journal of Consulting & Clinical Psychology 1993; 61:296-305.
Wilfley, D., Welch, S. S., Stein, R. I., Spurrell, E. B., Cohen, L. R., Saelens, B. E., Dounchis, J. Z., Frank, M. A., Wiseman, C. V., and Matt, G. E. A randomized comparison of group cognitivebehavioral therapy and group interpersonal psychotherapy for the treatment of overweight individuals with binge-eating disorder. Archives of General Psychiatry 2002; 25:713-721.
Wilhelm, K and Clarke, D. Eating Disorders from a Primary Care Perspective. Medical Journal of Australia, 168: 458-463 Whitehouse AM, Cooper PJ, Vize CV, Hill C, Vogel L. Prevalence of eating disorders in three Cambridge general practices: hidden and conspicuous morbidity. British Journal of General Practice 1992; 42(355):57-60.
Willi J, Hagemann R. Long-term course of anorexia nervosa Schweiz Med Wochenschr. 1976 Oct 23;106(43):1459-65. German.
Williams P, Hand D, Tarnopolsky A The problem of screening for uncommon disorders - a comment on the Eating Attitudes Test. Psychol Med. 1982 May;12(2):431-4.
Williamson, D.A.; Thaw, J.M.; Varnado-Sullivan, P.J. Cost-effectiveness analysis of a hospital- based cognitive-behavioural treatment program for eating disorders. Behaviour Therapy, 2001; 32: 459-477
Wilson, G. T., Eldredge, K. L., Smith, D., and Niles, B. Cognitive behavioral treatment with and without response prevention for bulimia. Behaviour Research and Therapy 1991; 29:575- 583.
Wilson, G. T., Fairburn, C. C., Agras, W. S., Walsh, B. T., and Kraemer, H. Cognitive- behavioral therapy for bulimia nervosa: time course and mechanisms of change. Journal of Consulting & Clinical Psychology 2002; 70:267-274.
Wilson, G. T., Loeb, K. L., Walsh, B. T., Labouvie, E., Petkova, E., Liu, X., and Waternaux, C. Psychological versus pharmacological treatments of bulimia nervosa: Predictors and
234 CONFIDENTIAL Second draft for consultation September 2003 processes of change. Journal of Consulting & Clinical Psychology 1999;67:451-459.Wilson et al., 1986 Treatment of bulimia nervosa: when CBT fails. Behav Res Ther.
Wolf, E. M. and Crowther, J. H. An evaluation of behavioral and cognitive-behavioral group interventions for the treatment of bulimia nervosa in women. International Journal of Eating Disorders 1992; 11:3-15.
Wolk, S. L. and Devlin, M. J. Stage of change as a predictor of response to psychotherapy for bulimia nervosa. International Journal of Eating Disorders 2001; 30:96-100.
Zipfel S, Löwe B, Reas D L, et al. Long-Term Prognosis in Anorexia Nervosa: Lessons From a 21 Year Follow-Up Study. The Lancet 2000; 355:712-722.
Zipfel S, Seibel MJ, Lowe B, Beumont PJ, Kasperk C, Herzog W: Osteoporosis in eating disorders: a follow-up study of patients with anorexia and bulimia. Journal of Clinical Endocrinology and Metabolism 2001; 86:5227-5233.
235 CONFIDENTIAL Second draft for consultation September 2003
Abbreviations and glossary of terms Abbreviations
A&E Accident and emergency AGREE Appraisal of Guidelines Research and Evaluation APA American Psychiatric Association
BED Binge eating disorder BMI Body mass index BPS British Psychological Society
CAMHS Child and adolescent mental health services CAT Cognitive analytic therapy CBT Cognitive behaviour therapy CEBMH Centre for Evidence Based Mental Health CEFAHP Clinical Effectiveness Forum for the Allied Health Professions CEMH Centre for Economics in Mental Health CHAI Commission for Health Care, Audit and Improvement CI Confidence interval CMR Crude mortality rate CORE British Psychological Society Centre for Outcomes Research and Effectiveness COT College of Occupational Therapists CRU Royal College of Psychiatrists College Research Unit
DHEA Dehyroepiandosterone DSM-IV Diagnostic and Statistical Manual of Mental Disorders, Fourth edition
EAT Eating attitudes test EDE Eating disorders examination EDI Eating disorders inventory EDNOS Eating disorders not otherwise specified EDSIG Eating disorders special interest group EEG Electroencephalography ERP Exposure with response prevention
GDG Guideline development group GSH Guided self help
ICD10 International Classification of Diseases, 10th Edition IoP Institute of Psychiatry IPT Interpersonal Psychotherapy
236 CONFIDENTIAL Second draft for consultation September 2003
MANOVA Multivariate analysis of variance MAOI Monoamine-oxidase inhibitors
N Number of studies n Number of participants NCCMH National Collaborating Centre for Mental Health NG Nasogastric NICAPS National Inpatient Child and Adolescent Psychiatry Study NHS National Health Service NICE National Institute for Clinical Excellence NNT Number needed to treat
OCD Obsessive compulsive disorder OR Odds ratio
PCTs Primary Care Trusts PEG Percutaneous endoscopic gastrostomy PSH Pure self help
RCGP Royal College of General Practitioners RCN Royal College of Nursing RCT Randomised controlled trials RCPsych Royal College of Psychiatrists RPS Royal Pharmaceutical Society RR Relative risk, risk ratio
SCIE Social Care Institute of Excellence SMR Standardised Mortality Rate SSRI Selective serotonin reuptake inhibitors
TPN Total parenteral nutrition
WHO World Health Organization
237 CONFIDENTIAL Second draft for consultation September 2003 12 Glossary
Adherence: The behaviour of taking medicine according to treatment dosage and schedule as intended by the prescriber. In this guideline, the term ‘adherence’ is used in preference to ‘compliance’, but is not synonymous with ‘concordance’, which has a number of meanings.
Affective disorder: [to be added ss 2.5.5]
Anorexia nervosa: A syndrome in which the individual maintains a low weight as a result of a preoccupation with body weight, construed either as a fear of fatness or pursuit of thinness. Weight is maintained at least 15% below that expected, that is, Body Mass Index (calculated as weight in kilograms divided by height in metres squared) is below 17.5.
Atypical eating disorder: This term is used to denote eating disorders of clinical severity that do not meet the diagnostic criteria for AN or BN. The equivalent American term is “eating disorder not otherwise specified.”
Behaviour therapy (BT): Behavioural treatments that include many features of cognitive behaviour therapy – bulimia nervosa (CBT-BN) but have little or no emphasis on the direct modification of ways of thinking.
Binge eating disorder (BED): A syndrome in which an individual experiences repeated uncontrolled episodes of overeating but does not use extreme compensatory weight-control behaviours. It is a provisional new eating disorder diagnosis in DSM-IV.
Bulimia nervosa: A syndrome characterised by recurrent episodes of binge eating and by compensatory behaviour (vomiting, purging, fasting or exercising) in order to prevent weight gain. Binge eating is accompanied by a subjective feeling of loss of control over eating. This is a normal weight syndrome in which the body mass index (BMI) is maintained above 17.5 kg/m2.
Cognitive analytic therapy (CAT): [to be added ss 2.7.2]
Cognitive behavioural therapy (CBT): A psychological intervention that is designed to enable people to establish links between their thoughts, feelings or actions and their current or past symptoms and to re-evaluate their perceptions, beliefs or reasoning about the target symptoms. The intervention should involve at least one of the following: (1) monitoring thoughts, feelings or behaviour with respect to the symptom; (2) being helped to use alternative ways of coping with the target symptom; (3) reducing stress.
Cognitive behaviour therapy for Bulimia Nervosa (CBT-BN): A specific form of CBT devised for patients with bulimia nervosa. It typically involves
238 CONFIDENTIAL Second draft for consultation September 2003 16-20 hour-long one-to-one treatment sessions over four-to-five months. Its focus is not only on helping patients change their eating habits but also on addressing the ways of thinking (most especially the over-evaluation of shape and weight) that maintain them. It has been described in several treatment manuals, the one used in most of the more recent treatment trials being that by Fairburn and colleagues (Fairburn et al., 1993).
Confidence interval: The range within which the ‘true‘ values (e.g., size of effect of an intervention) are expected to lie with a given degree of certainty (e.g., 95% or 99%). Confidence intervals represent the probability of random errors, but not systematic errors – or bias.
Controlled Trial – An experiment in which investigators allocate eligible people into groups to receive or not to receive one or more interventions that are being compared.
Costing study: The simplest economic study, measuring only the costs of given interventions. It does not provide answers to efficiency questions.
Cost-benefit analysis: A type of full economic evaluation that compares alternatives in which the costs and consequences vary. Both costs and benefits are measured in monetary units. If benefits exceed costs, the evaluation would be a basis for recommending the treatment. It can address the question of whether a treatment or policy is socially worthwhile in the broadest sense.
Cost-consequence analysis: An analysis in which both outcomes and costs of alternative treatments or policy options are described. However, multiple outcomes are measured and there is no attempt to reduce everything to a single ratio.
Cost-effectiveness analysis: A type of full economic evaluation that compares competing alternatives of which the costs and consequences vary. The outcomes are measured in the same non-monetary (natural) unit. It expresses the result in the form of an incremental (or average or marginal) cost- effectiveness ratio.
Cost-minimisation analysis: A type of full economic evaluation where the costs of equally effective alternative treatments are compared. The aim is to find the least costly alternative.
Cost-of-illness/economic burden studies: An economic analysis of the total costs incurred by a society due to a specific disease.
Costs (direct): The costs of all the goods, services and other resources that are consumed in the provision of a health intervention. They can be medical or non-medical.
239 CONFIDENTIAL Second draft for consultation September 2003 Costs (indirect): The lost productivity suffered by the national economy as a result of an employee’s absence from the workplace through illness, decreased efficiency or premature death.
Cost-utility analysis: A type of full economic evaluations that compares competing alternatives of which the costs and consequences vary. It measures and values the impact of a treatment or policy alternative in utility units (see QALY). The result is expressed in the form of a cost-utility ratio.
Decision analysis: An explicit, quantitative, systematic approach to decision- making under conditions of uncertainty, in which the probability of each possible event, along with the consequences of those events, is explicitly stated.
Dialectical behaviour therapy (DBT): A multifaceted and intensive psychological treatment designed for patients with borderline personality disorder (Linehan, 1993a, 1993b). A simplified and abbreviated form of the treatment has been developed for patients with bulimia nervosa or binge eating disorder (Telch et al., 2000, 2001). It primarily focuses on enhancing patients’ emotion regulation skills. It involves 20 two-hour group sessions once a week. Details are provided by Wiser and Telch (1999).
Dietary counselling: Also termed “nutritional counselling”. A form of treatment in which the primary goal is the modification of what the patient eats as well as relevant eating habits and attitudes. It is not a well-defined intervention and is practised in a variety of ways. It is usually implemented by dieticians.
Drop out: A term no longer used to indicate leaving a study before its completion (the term ‘leaving the study early’ is now preferred).
Economic evaluation: Technique developed to assess both costs and consequences of alternative health strategies and to provide a decision- making framework.
Eating disorders not otherwise specified (EDNOS): Eating disorders which closely resemble anorexia nervosa and bulimia nervosa, but are considered atypical as they do not meet the precise diagnostic criteria for these conditions.
Effectiveness: The extent to which a specific intervention, when used under ordinary circumstances, does what it is intended to do. Clinical trials that assess effectiveness are sometimes called management trials.
Efficacy: The extent to which an intervention produces a beneficial result under ideal conditions. Clinical trials that assess efficacy are sometimes called explanatory trials and are restricted to participants who fully co-operate. The
240 CONFIDENTIAL Second draft for consultation September 2003 randomised controlled trial is the accepted ‘gold standard’ for evaluating the efficacy of an intervention.
Exposure with Response Prevention (ERP): This form of treatment was originally based on an ‘anxiety reduction’ model of bulimia nervosa according to which self-induced vomiting negatively reinforces binge eating by removing fears of secondary weight gain (Rosen & Leitenberg, 1982). The model on which it is based has evolved over the years (Carter & Bulik, 1994) and various versions of the treatment have been tested (e.g., Rosen & Leitenberg, 1982; Jansen et al., 1989; Schmidt & Marks, 1989). Each involves repeated exposure to cues that precede binge eating or purging with the response (binge eating or purging) being prevented. The treatment is typically combined with elements of cognitive behaviour therapy. Treatment sessions that involve in vivo exposure may last up to three hours. ERP has been describes as ‘time-intensive, expensive and logistically complicated’ (Bulik et al., 1998, page 620).
Family interventions: Family sessions with a treatment function based on systemic, cognitive behavioural or psychoanalytic principles, which may include psychoeducational, problem-solving and crisis management work and specific interventions with the identified patient. With eating disorders, the focus is on the eating disorder and how this impacts family relationships, emphasising in the early stages of treatment the necessity for parents to take a central role in supporting their child’s efforts to eat.
Feeding against the will of the patient: Using any method of feeding that requires that the individual be restrained to allow the feeding to take place.
Focal Supportive Psychotherapy: A collective term used in this Guideline for the various types of supportive psychological treatment that have been evaluated in the treatment of patients with eating disorders. These have varied in nature but they have in common their focus on supporting patients in their attempts to address difficulties in their life. Some have included limited emphasis on changing eating habits. These treatments should not be confused with more general forms of supportive psychotherapy or with counselling.
Forest plot: A graphical display of results from individual studies on a common scale, allowing visual comparison of trial results and examination of the degree of heterogeneity between studies.
Guided self-help (GSH): A self-help programme for bulimia nervosa in which a clinical professional provides support and encouragement.
Guideline recommendation: A systematically developed statement that is derived from the best available research evidence, using predetermined and systematic methods to identify and evaluate evidence relating to the specific condition in question.
241 CONFIDENTIAL Second draft for consultation September 2003 Insulin purging: The omission or intentional under-dosing with insulin, sometimes done by patients with bulimia nervosa and diabetes.
Interpersonal psychotherapy (IPT): A specific form of focal psychotherapy that is designed to help patients identify and address current interpersonal problems. It was originally developed for the treatment of depression (Klerman et al., 1984) but has been adapted for the treatment of bulimia nervosa (IPT-BN; Fairburn, 1997). In IPT-BN there is no emphasis on directly modifying eating habits; rather, it is expected that they will change as interpersonal functioning improves. IPT-BN involves the same number and pattern of treatment sessions as CBT– BN.
Meta-analysis: The use of statistical techniques in a systematic review to integrate the results of the included studies. Also used to refer to systematic reviews that use meta-analysis.
Number of people leaving the study early: For the purposes of the guideline, the number of people leaving the study early due to any reason is taken as a proxy for treatment acceptability, whereas the number of people leaving the study early due to adverse events is taken as a proxy for treatment tolerability. An exception to this is when the comparison group is a wait-list control, in which case this assumption is not made.
Number needed to harm: The number of people (calculated statistically) who need to be treated to cause one bad outcome. The lower the number needed to harm, the higher the likelihood of harm.
Number needed to treat: The number of people who need to be treated to prevent one bad outcome (i.e., a good outcome). It is the inverse of the risk difference.
Nutritional counselling: See Dietary counselling
Odds ratio: a measure of the relative benefit of the experimental treatment that can be obtained by dividing the experimental odds by the control odds.
Patient: The terms ‘patient’ or ‘person with eating disorder, anorexia nervosa, bulimia nervosa or binge eating disorder, etc.’, are used in this guideline to identify the person presently or formerly with the condition and/or receiving services in the present or past.
Psychodynamic psychotherapy: Regular individual therapy sessions with a trained psychotherapist, or a therapist under supervision, based on a psychodynamic or psychoanalytic model, which use a variety of strategies, including exploratory insight-oriented, supportive or directive activity, applied flexibly, working with transference, but with the therapists using a less strict technique than that used in psychoanalysis.
242 CONFIDENTIAL Second draft for consultation September 2003 Psychological intervention – to be added
Pure self help (PSH): A self-help programme for bulimia nervosa that patients follow on their own without the support of a professional.
Quality-adjusted life years (QALY): A form of utility measure, calculated by estimating the total life-years gained from a treatment and weighting each year with a quality-of-life score in that year.
Randomisation: Method used to generate a random allocation sequence, such as using tables of random numbers or computer-generated random sequences. The method of randomisation should be distinguished from concealment of allocation, because if the latter is inadequate selection bias may occur despite the use of randomisation. For instance, a list of random numbers may be used to randomise participants, but if the list were open to the individuals responsible for recruiting and allocating participants, those individuals could influence the allocation process, either knowingly or unknowingly.
Randomised controlled trial (RCT): Also termed ‘randomised clinical trial’. An experiment in which investigators randomly allocate eligible people into groups to receive or not to receive one or more interventions that are being compared. The results are assessed by comparing outcomes in the different groups. Through randomisation, the groups should be similar in all aspects apart from the treatment they receive during the study.
Refeeding: increasing nutritional intake and restoring weight to within a normal range in the treatment of anorexia nervosa
Relative risk: Also known as risk ratio; the ratio of risk in the intervention group to the risk in the control group. The risk (proportion, probability or rate) is the ratio of people with an event in a group to the total in the group. A relative risk (RR) of 1 indicates no difference between comparison groups. For undesirable outcomes, an RR that is less than 1 indicates that the intervention was effective in reducing the risk of that outcome.
Selective serotonin reuptake inhibitors (SSRIs): Medicines that inhibit the reuptake of serotonin into the presynaptic neurone thus increasing neurotransmission. Although they ‘selectively’ inhibit serotonin reptake, they are not serotonin specific. Some of the drugs in this class also inhibit the reuptake of noradrenaline and/or dopamine. As a class, these drugs are associated with less anticholinergic side effects and are less likely to cause postural hypotension or sedation.
Self-help: This involves following a self-help programme for bulimia nervosa, either with support and encouragement from a clinical professional (guided self-help) or on one’s own (pure self-help). The self-help programmes studied
243 CONFIDENTIAL Second draft for consultation September 2003 to date have taken the form of self-help books (Cooper, 1995; Fairburn, 1995; Schmidt & Treasure, 1995).
Sensitivity analysis: Sensitivity analysis is a technique used in economic analysis or decision-making to allow for uncertainty by testing whether plausible changes in the values of the main variables affect the results of the analysis.
Standard dosage: The recommended dosage range listed in the British National Formulary; this normally reflects the information contained in the manufacturers’ Summary of Product Characteristics as well as advice from an external panel of experts.
Stepped-care model: A sequence of treatment options to offer simpler and less expensive interventions first and more complex and expensive interventions if the patient has not benefited, based on locally-agreed protocols.
Weighted mean difference: A method of meta-analysis used to combine measures on continuous scales (such as weight), where the mean, standard deviation and sample size in each group are known. The weight given to each study (e.g., how much influence each study has on the overall results of the meta-analysis) is determined by the precision of its estimate of effect and, in the statistical software used by the NCCMH, is equal to the inverse of the variance. This method assumes that all of the trials have measured the outcome on the same scale.
244